commit 432d646588ca6c2e8d0f126e88bc25131bf26b8a
Author: CI <toni@data-intuitive.com>
Date:   Wed Apr 8 12:30:26 2026 +0000

    Build branch openpipeline_composed/set-public-scope with version set-public-scope to openpipeline_composed on branch set-public-scope (5068271)
    
    Build pipeline: openpipelines-bio.openpipeline-composed.main-vnhhz
    
    Source commit: https://github.com/openpipelines-bio/openpipeline_composed/commit/50682718d61f38677e46dde857fb948846ec6cf6
    
    Source message: Merge branch 'set-public-scope' of github.com:openpipelines-bio/openpipeline_composed into set-public-scope

diff --git a/.gitignore b/.gitignore
new file mode 100644
index 0000000..13efa86
--- /dev/null
+++ b/.gitignore
@@ -0,0 +1,26 @@
+# IDEs and editors
+/.idea
+.project
+.classpath
+*.launch
+.settings/
+.vscode
+
+# Temp
+gitignore
+test_results
+
+# System Files
+.DS_Store
+Thumbs.db
+
+# Nextflow
+work
+.nextflow*
+trace-*.txt
+
+# viash
+/resources_test/
+
+# pycache
+*__pycache__*
diff --git a/CHANGELOG.md b/CHANGELOG.md
new file mode 100644
index 0000000..3a549e5
--- /dev/null
+++ b/CHANGELOG.md
@@ -0,0 +1,19 @@
+# openpipeline_composed x.x.x
+
+## MINOR CHANGES
+
+* `workflows/single_cell/process_integrate_annotate`: Set scope to `private` (PR #6).
+
+* Bump `openpipeline` dependency version to `v4.0.4` (PR #9).
+
+* Bump `viash` version to `0.9.7` (PR #10).
+
+# openpipeline_composed 0.1.1
+
+## MINOR CHANGES
+
+* Add a README (PR #4).
+
+# openpipeline_composed 0.1.0
+
+Initial release containing a single-cell meta-workflow to process single cell omics samples, perform batch integration and/or label projection.
\ No newline at end of file
diff --git a/LICENSE b/LICENSE
new file mode 100644
index 0000000..c12f874
--- /dev/null
+++ b/LICENSE
@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025 openpipelines-bio
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.
diff --git a/README.md b/README.md
new file mode 100644
index 0000000..60eb989
--- /dev/null
+++ b/README.md
@@ -0,0 +1,42 @@
+# OpenPipeline Composed
+
+OpenPipeline Composed provides a comprehensive meta-workflow that combines multiple stand-alone workflows from the [OpenPipeline](https://github.com/openpipelines-bio/openpipeline/) package. The meta-workflow combines sample processing, batch integration, and cell type annotation into a unified pipeline for single-cell multi-omics data analysis.
+
+[![ViashHub](https://img.shields.io/badge/ViashHub-openpipeline_composed-7a4baa.svg)](https://www.viash-hub.com/packages/openpipeline_composed)
+[![GitHub](https://img.shields.io/badge/GitHub-openpipelines--bio%2Fopenpipeline_composed-blue.svg)](https://github.com/openpipelines-bio/openpipeline_composed)
+[![GitHub License](https://img.shields.io/github/license/openpipelines-bio/openpipeline_composed.svg)](https://github.com/openpipelines-bio/openpipeline_composed/blob/main/LICENSE)
+[![GitHub Issues](https://img.shields.io/github/issues/openpipelines-bio/openpipeline_composed.svg)](https://github.com/openpipelines-bio/openpipeline_composed/issues)
+[![Viash version](https://img.shields.io/badge/Viash-v0.9.4-blue.svg)](https://viash.io)
+
+## Overview
+
+The sole purpose of this package is to provide a meta-workflow that orchestrates and combines various stand-alone workflows from the [OpenPipeline](https://github.com/openpipelines-bio/openpipeline/) package. By integrating multiple processing steps into a single workflow, it enables seamless processing from raw data to fully annotated, integrated datasets suitable for downstream analysis and atlas generation.
+
+## Functionality
+
+The meta-workflow combines three core OpenPipeline workflows:
+- [**Sample Processing**](https://www.viash-hub.com/packages/openpipeline/latest/components/workflows/multiomics/process_samples): Initial quality control, filtering, and preprocessing
+- [**Batch Integration**](https://www.viash-hub.com/packages/openpipeline/latest/components?search=workflows%2Fintegration): Integration using **Harmony** or **scVI** methods
+- [**Cell Type Annotation**](https://www.viash-hub.com/packages/openpipeline/latest/components?search=workflows%2Fannotation): Annotation using **scANVI** or **CellTypist** methods
+
+## Key Features
+
+- 🔄 **End-to-End Processing**: Complete pipeline from raw data to annotated results
+- 📊 **Atlas Generation**: Create comprehensive atlases from multiple datasets and sources
+- 🔬 **Multi-Modal Support**: Process RNA-seq, ATAC-seq, protein, and spatial data
+- 🎯 **Method Flexibility**: Choose from multiple integration and annotation approaches
+- 🧬 **Reference Integration**: Leverage existing reference datasets for annotation
+
+## Execution via CLI or Seqera Cloud
+
+The openpipeline_composed package is available via [Viash Hub](https://www.viash-hub.com/packages/openpipeline_composed/latest/), where you can receive instructions on how to run the end-to-end workflow as well as individual subworkflows or components.
+
+It's possible to run the workflow directly from Seqera Cloud. The necessary Nextflow schema files have been built and provided with the workflows in order to use the form-based input. However, Seqera Cloud can not deal with multiple-value parameters for batch processing of multiple samples. Therefore, it's better to use Viash Hub also here for launching the workflow on Seqera Cloud.
+
+* Navigate to the [Viash Hub package page](https://www.viash-hub.com/packages/openpipeline_composed/latest/), select the workflow you want to launch and click the `launch` button.
+* Select the execution environment of choice (e.g. `Seqera Cloud`, `Nextflow` or `Executable`)
+* Fill in the form with the required parameters and launch the workflow.
+
+## Support
+
+For issues specific to the composed meta-workflow, please use the [GitHub issues tracker](https://github.com/openpipelines-bio/openpipeline_composed/issues). For general OpenPipeline questions, refer to the main [OpenPipeline documentation](https://openpipelines.bio/).
diff --git a/_viash.yaml b/_viash.yaml
new file mode 100644
index 0000000..f1242b3
--- /dev/null
+++ b/_viash.yaml
@@ -0,0 +1,32 @@
+viash_version: 0.9.7
+source: src
+target: target
+name: openpipeline_composed
+organization: vsh
+links:
+  repository: https://github.com/openpipelines-bio/openpipeline_composed
+  docker_registry: ghcr.io
+repositories:
+  - name: openpipeline
+    repo: openpipeline
+    type: vsh
+    tag: v4.0.4
+  - name: openpipeline_qc
+    repo: openpipeline_qc
+    type: vsh
+    tag: v0.2.2
+  - name: biobox
+    repo: biobox
+    type: vsh
+    tag: v0.4.2
+info:
+  test_resources:
+    - type: s3
+      path: s3://openpipelines-bio/openpipeline_incubator/resources_test
+      dest: resources_test
+config_mods: |
+  .requirements.commands := ['ps']
+  .runners[.type == 'nextflow'].directives.tag := '$id'
+  .resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}
+  .runners[.type == 'nextflow'].config.script := 'includeConfig("nextflow_labels.config")'
+version: set-public-scope
diff --git a/main.nf b/main.nf
new file mode 100644
index 0000000..e69de29
diff --git a/nextflow.config b/nextflow.config
new file mode 100644
index 0000000..e69de29
diff --git a/resources_test_scripts/10x_5k_anticmv.sh b/resources_test_scripts/10x_5k_anticmv.sh
new file mode 100644
index 0000000..aee5c20
--- /dev/null
+++ b/resources_test_scripts/10x_5k_anticmv.sh
@@ -0,0 +1,170 @@
+#!/bin/bash
+
+set -eo pipefail
+
+
+# ensure that the command below is run from the root of the repository
+REPO_ROOT=$(git rev-parse --show-toplevel)
+cd "$REPO_ROOT"
+
+# settings
+ID=10x_5k_anticmv
+OUT=resources_test/$ID
+
+# create raw directory
+raw_dir="$OUT/raw"
+mkdir -p "$raw_dir"
+
+# Check whether seqkit is available
+if ! command -v seqkit &> /dev/null; then
+    echo "This script requires seqkit. Please make sure the binary is added to your PATH."
+    exit 1
+fi
+
+# dataset page:
+# https://www.10xgenomics.com/resources/datasets/integrated-gex-totalseqc-and-tcr-analysis-of-connect-generated-library-from-5k-cmv-t-cells-2-standard
+
+# check whether reference is available
+reference_dir="resources_test/reference_gencodev41_chr1/"
+genome_tar="$reference_dir/reference_cellranger.tar.gz"
+if [[ ! -f "$genome_tar" ]]; then
+    echo "$genome_tar does not exist. Please create the reference genome first"
+    exit 1
+fi
+
+# download and untar source fastq files
+tar_dir="$HOME/.cache/openpipeline/5k_human_antiCMV_T_TBNK_connect_Multiplex"
+if [[ ! -d "$tar_dir" ]]; then
+    mkdir -p "$tar_dir"
+
+    # download fastqs and untar
+    wget "https://s3-us-west-2.amazonaws.com/10x.files/samples/cell-vdj/6.1.2/5k_human_antiCMV_T_TBNK_connect_Multiplex/5k_human_antiCMV_T_TBNK_connect_Multiplex_fastqs.tar" -O "$tar_dir.tar"
+    tar -xvf "$tar_dir.tar" -C "$tar_dir" --strip-components=1
+    rm "$tar_dir.tar"
+fi
+
+function seqkit_head {
+  input="$1"
+  output="$2"
+  if [[ ! -f "$output" ]]; then
+    echo "> Processing `basename $input`"
+    seqkit head -n 200000 "$input" | gzip > "$output"
+  fi
+}
+
+orig_sample_id="5k_human_antiCMV_T_TBNK_connect"
+
+seqkit_head "$tar_dir/gex_1/${orig_sample_id}_GEX_1_S1_L001_R1_001.fastq.gz" "$raw_dir/${orig_sample_id}_GEX_1_subset_S1_L001_R1_001.fastq.gz"
+seqkit_head "$tar_dir/gex_1/${orig_sample_id}_GEX_1_S1_L001_R2_001.fastq.gz" "$raw_dir/${orig_sample_id}_GEX_1_subset_S1_L001_R2_001.fastq.gz"
+
+seqkit_head "$tar_dir/ab/${orig_sample_id}_AB_S2_L004_R1_001.fastq.gz" "$raw_dir/${orig_sample_id}_AB_subset_S2_L004_R1_001.fastq.gz"
+seqkit_head "$tar_dir/ab/${orig_sample_id}_AB_S2_L004_R2_001.fastq.gz" "$raw_dir/${orig_sample_id}_AB_subset_S2_L004_R2_001.fastq.gz"
+
+seqkit_head "$tar_dir/vdj/${orig_sample_id}_VDJ_S1_L001_R1_001.fastq.gz" "$raw_dir/${orig_sample_id}_VDJ_subset_S1_L001_R1_001.fastq.gz"
+seqkit_head "$tar_dir/vdj/${orig_sample_id}_VDJ_S1_L001_R2_001.fastq.gz" "$raw_dir/${orig_sample_id}_VDJ_subset_S1_L001_R2_001.fastq.gz"
+
+# download immune panel fasta if needed
+feature_reference="$raw_dir/feature_reference.csv"
+if [[ ! -f "$feature_reference" ]]; then
+  wget "https://cf.10xgenomics.com/samples/cell-vdj/6.1.2/5k_human_antiCMV_T_TBNK_connect_Multiplex/5k_human_antiCMV_T_TBNK_connect_Multiplex_count_feature_reference.csv" -O "$feature_reference"
+fi
+
+# download vdj reference if needed
+vdj_ref="$raw_dir/refdata-cellranger-vdj-GRCh38-alts-ensembl-7.0.0.tar.gz"
+if [[ ! -f "$vdj_ref" ]]; then
+  wget "https://cf.10xgenomics.com/supp/cell-vdj/refdata-cellranger-vdj-GRCh38-alts-ensembl-7.0.0.tar.gz" -O "$vdj_ref"
+fi
+
+
+# Run mapping pipeline
+cat > /tmp/params.yaml << HERE
+param_list:
+- id: "$ID"
+  input: "$raw_dir"
+  library_id:
+    - "${orig_sample_id}_GEX_1_subset"
+    - "${orig_sample_id}_AB_subset"
+    - "${orig_sample_id}_VDJ_subset"
+  library_type:
+    - "Gene Expression"
+    - "Antibody Capture"
+    - "VDJ"
+
+gex_reference: "$genome_tar"
+vdj_reference: "$vdj_ref"
+feature_reference: "$feature_reference"
+HERE
+
+nextflow \
+  run https://packages.viash-hub.com/vsh/openpipeline \
+  -r v4.0.4 \
+  -main-script target/nextflow/mapping/cellranger_multi/main.nf \
+  -resume \
+  --publish_dir "${OUT}_v10/processed" \
+  -profile docker,mount_temp \
+  -params-file /tmp/params.yaml \
+  -c ./src/configs/labels_ci.config 
+
+# Convert to h5mu
+cat > /tmp/params.yaml << HERE
+id: "$orig_sample_id"
+input: "$OUT/processed/10x_5k_anticmv.cellranger_multi.output"
+publish_dir: "$OUT/"
+output: "*.h5mu"
+HERE
+
+nextflow \
+  run https://packages.viash-hub.com/vsh/openpipeline \
+  -r v4.0.4 \
+  -main-script target/nextflow/convert/from_cellranger_multi_to_h5mu/main.nf \
+  -resume \
+  -profile docker,mount_temp \
+  -params-file /tmp/params.yaml \
+  -c ./src/configs/labels_ci.config
+
+mv "$OUT/0.h5mu" "$OUT/${orig_sample_id}.h5mu"
+
+
+# run qc workflow
+cat > /tmp/params.yaml << HERE
+id: "$ID"
+input: "$OUT/$orig_sample_id.h5mu"
+var_name_mitochondrial_genes: mitochondrial
+var_name_ribosomal_genes: ribosomal
+publish_dir: "$OUT/"
+output: "${orig_sample_id}_qc.h5mu"
+HERE
+
+nextflow \
+  run https://packages.viash-hub.com/vsh/openpipeline \
+  -r v4.0.4 \
+  -main-script target/nextflow/workflows/qc/qc/main.nf \
+  -resume \
+  -profile docker,mount_temp \
+  -params-file /tmp/params.yaml \
+  -c ./src/configs/labels_ci.config
+
+
+# Run full pipeline
+cat > /tmp/params.yaml << HERE
+id: "$ID"
+input: "$OUT/${orig_sample_id}_qc.h5mu"
+publish_dir: "$OUT/"
+output: "${orig_sample_id}_mms.h5mu"
+HERE
+
+nextflow \
+  run https://packages.viash-hub.com/vsh/openpipeline \
+  -r v4.0.4 \
+  -main-script target/nextflow/workflows/multiomics/process_samples/main.nf \
+  -resume \
+  -profile docker,mount_temp \
+  -params-file /tmp/params.yaml \
+  -c ./src/configs/labels_ci.config
+
+aws s3 sync \
+  "$OUT" \
+  s3://openpipelines-bio/openpipeline_incubator/resources_test/"$ID" \
+  --exclude "*.yaml" \
+  --delete \
+  --dryrun
diff --git a/resources_test_scripts/annotation_test_data.sh b/resources_test_scripts/annotation_test_data.sh
new file mode 100644
index 0000000..e989564
--- /dev/null
+++ b/resources_test_scripts/annotation_test_data.sh
@@ -0,0 +1,166 @@
+#!/bin/bash
+
+set -eo pipefail
+
+# get the root of the directory
+REPO_ROOT=$(git rev-parse --show-toplevel)
+
+# ensure that the command below is run from the root of the repository
+cd "$REPO_ROOT"
+
+ID=annotation_test_data
+OUT=resources_test/$ID/
+
+# ideally, this would be a versioned pipeline run
+[ -d "$OUT" ] || mkdir -p "$OUT"
+
+# Download Tabula Sapiens Blood reference h5ad from https://doi.org/10.5281/zenodo.7587774
+wget "https://zenodo.org/record/7587774/files/TS_Blood_filtered.h5ad?download=1" -O "${OUT}/tmp_TS_Blood_filtered.h5ad"
+
+# Download Tabula Sapiens Blood pretrained model from https://doi.org/10.5281/zenodo.7580707
+wget "https://zenodo.org/record/7580707/files/pretrained_models_Blood_ts.tar.gz?download=1" -O "${OUT}/tmp_pretrained_models_Blood_ts.tar.gz"
+
+# Download PopV specific CL ontology files - needed for OnClass
+# OUT_ONTOLOGY="${OUT}/ontology"
+# [ -d "$OUT_ONTOLOGY" ] || mkdir -p "$OUT_ONTOLOGY"
+# wget https://raw.githubusercontent.com/czbiohub/PopV/main/ontology/cl.obo \
+# -O "${OUT_ONTOLOGY}/cl.obo"
+# wget https://raw.githubusercontent.com/czbiohub/PopV/main/ontology/cl.ontology \
+# -O "${OUT_ONTOLOGY}/cl.ontology"
+# wget https://raw.githubusercontent.com/czbiohub/PopV/main/ontology/cl.ontology.nlp.emb \
+# -O "${OUT_ONTOLOGY}/cl.ontology.nlp.emb"
+
+
+# Process Tabula Sapiens Blood reference h5ad
+# (Select one individual and 100 cells per cell type)
+# normalize and log1p transform data
+# Add treatment and disease columns
+python <<HEREDOC
+import anndata as ad
+import scanpy as sc
+import numpy as np
+
+# Read in data
+ref_adata = ad.read_h5ad("${OUT}/tmp_TS_Blood_filtered.h5ad")
+sub_ref_adata = ref_adata[ref_adata.obs["donor_assay"] == "TSP14_10x 3' v3"] 
+n=100
+s=sub_ref_adata.obs.groupby('cell_ontology_class').cell_ontology_class.transform('count')
+sub_ref_adata_final = sub_ref_adata[sub_ref_adata.obs[s>=n].groupby('cell_ontology_class').head(n).index]
+
+# Normalize and log1p transform data
+data_for_scanpy = ad.AnnData(X=sub_ref_adata_final.X)
+sc.pp.normalize_total(data_for_scanpy, target_sum=10000)
+sc.pp.log1p(
+    data_for_scanpy,
+    base=None,
+    layer=None,
+    copy=False,
+)
+sub_ref_adata_final.layers["log_normalized"] = data_for_scanpy.X
+
+# Add treatment and disease columns
+n_cells = sub_ref_adata_final.n_obs
+treatment = np.random.choice(["ctrl", "stim"], size=n_cells, p=[0.5, 0.5])
+disease = np.random.choice(["healthy", "diseased"], size=n_cells, p=[0.5, 0.5])
+sub_ref_adata_final.obs["treatment"] = treatment
+sub_ref_adata_final.obs["disease"] = disease
+
+# Write out data
+sub_ref_adata_final.write("${OUT}/TS_Blood_filtered.h5ad", compression='gzip')
+HEREDOC
+
+
+echo "> Converting to h5mu"
+viash run src/convert/from_h5ad_to_h5mu/config.vsh.yaml --engine docker -- \
+    --input "${OUT}/TS_Blood_filtered.h5ad" \
+    --output "${OUT}/TS_Blood_filtered.h5mu" \
+    --modality "rna"
+
+rm "${OUT}/tmp_TS_Blood_filtered.h5ad"
+
+echo "> Downloading pretrained CellTypist model and sample test data"
+wget https://celltypist.cog.sanger.ac.uk/models/Pan_Immune_CellTypist/v2/Immune_All_Low.pkl \
+    -O "${OUT}/celltypist_model_Immune_All_Low.pkl"
+wget https://celltypist.cog.sanger.ac.uk/Notebook_demo_data/demo_2000_cells.h5ad \
+    -O "${OUT}/demo_2000_cells.h5ad"
+viash run src/convert/from_h5ad_to_h5mu/config.vsh.yaml --engine docker -- \
+    --input "${OUT}/demo_2000_cells.h5ad" \
+    --output "${OUT}/demo_2000_cells.h5mu" \
+    --modality "rna"
+
+
+echo "> Fetching OnClass data and models"
+OUT_ONTOLOGY="${OUT}/ontology"
+[ -d "$OUT_ONTOLOGY" ] || mkdir -p "$OUT_ONTOLOGY"
+wget https://figshare.com/ndownloader/files/28394466 -O "${OUT_ONTOLOGY}/OnClass_data_public_minimal.tar.gz"
+tar -xzvf "${OUT_ONTOLOGY}/OnClass_data_public_minimal.tar.gz" -C "${OUT_ONTOLOGY}" --strip-components=2
+rm "${OUT_ONTOLOGY}/allen.ontology"
+rm "${OUT_ONTOLOGY}/OnClass_data_public_minimal.tar.gz"
+
+wget https://figshare.com/ndownloader/files/28394541 -O "${OUT}/OnClass_models.tar.gz"
+tar -xzvf "${OUT}/OnClass_models.tar.gz" -C "${OUT}" --strip-components=1
+rm "${OUT}/OnClass_models.tar.gz"
+rm "${OUT}/tmp_pretrained_models_Blood_ts.tar.gz"
+
+find "${OUT}/Pretrained_model" ! -name "example_file_model*" -type f -exec rm -f {} +
+mv "${OUT}/Pretrained_model" "${OUT}/onclass_model"
+
+echo "> Creating simple SCVI model"
+viash run src/integrate/scvi/config.vsh.yaml --engine docker -- \
+    --input "${OUT}/TS_Blood_filtered.h5mu" \
+    --obs_batch "donor_id" \
+    --var_gene_names "ensemblid" \
+    --output "${OUT}/scvi_output.h5mu" \
+    --output_model "${OUT}/scvi_model" \
+    --max_epochs 5 \
+    --n_obs_min_count 10 \
+    --n_var_min_count 10
+
+echo "> Creating SCVI model with covariates"
+viash run src/integrate/scvi/config.vsh.yaml --engine docker -- \
+    --input "${OUT}/TS_Blood_filtered.h5mu" \
+    --obs_batch "donor_id" \
+    --var_gene_names "ensemblid" \
+    --obs_categorical_covariate "assay" \
+    --obs_categorical_covariate "donor_assay" \
+    --output "${OUT}/scvi_covariate_output.h5mu" \
+    --output_model "${OUT}/scvi_covariate_model" \
+    --max_epochs 5 \
+    --n_obs_min_count 10 \
+    --n_var_min_count 10
+
+echo "> Creating simple SCANVI model"
+viash run src/annotate/scanvi/config.vsh.yaml --engine docker -- \
+    --input "${OUT}/TS_Blood_filtered.h5mu" \
+    --var_gene_names "ensemblid" \
+    --obs_labels "cell_ontology_class" \
+    --scvi_model "${OUT}/scvi_model" \
+    --output "${OUT}/scanvi_output.h5mu" \
+    --output_model "${OUT}/scanvi_model" \
+    --max_epochs 5 
+
+echo "> Creating SCANVI model with covariates"
+viash run src/annotate/scanvi/config.vsh.yaml --engine docker -- \
+    --input "${OUT}/TS_Blood_filtered.h5mu" \
+    --var_gene_names "ensemblid" \
+    --obs_labels "cell_ontology_class" \
+    --scvi_model "${OUT}/scvi_covariate_model" \
+    --output "${OUT}/scanvi_covariate_output.h5mu" \
+    --output_model "${OUT}/scanvi_covariate_model" \
+    --max_epochs 5 
+
+rm "${OUT}/scanvi_output.h5mu"
+rm "${OUT}/scanvi_covariate_output.h5mu"
+rm "${OUT}/scvi_output.h5mu"
+rm "${OUT}/scvi_covariate_output.h5mu"
+rm -r "${OUT}/Pretrained_model/"
+
+echo "> Creating Pseudobulk Data for DGEA"
+viash run src/differential_expression/create_pseudobulk/config.vsh.yaml --engine docker -- \
+    --input "${OUT}/TS_Blood_filtered.h5mu" \
+    --obs_grouping "cell_type" \
+    --obs_sample_conditions "donor_id" \
+    --obs_sample_conditions "treatment" \
+    --obs_sample_conditions "disease" \
+    --min_num_cells_per_sample 5 \
+    --output "${OUT}/TS_Blood_filtered_pseudobulk.h5mu" 
diff --git a/resources_test_scripts/pbmc_1k_protein_v3.sh b/resources_test_scripts/pbmc_1k_protein_v3.sh
new file mode 100644
index 0000000..98ffd0f
--- /dev/null
+++ b/resources_test_scripts/pbmc_1k_protein_v3.sh
@@ -0,0 +1,151 @@
+#!/bin/bash
+
+set -eo pipefail
+
+# get the root of the directory
+REPO_ROOT=$(git rev-parse --show-toplevel)
+
+# ensure that the command below is run from the root of the repository
+cd "$REPO_ROOT"
+
+ID=pbmc_1k_protein_v3
+OUT=resources_test/$ID/$ID
+DIR=$(dirname "$OUT")
+
+# ideally, this would be a versioned pipeline run
+[ -d "$DIR" ] || mkdir -p "$DIR"
+
+# dataset page:
+# https://www.10xgenomics.com/resources/datasets/1-k-pbm-cs-from-a-healthy-donor-gene-expression-and-cell-surface-protein-3-standard-3-0-0
+
+# download metrics summary
+wget https://cf.10xgenomics.com/samples/cell-exp/3.0.0/pbmc_1k_protein_v3/pbmc_1k_protein_v3_metrics_summary.csv \
+  -O "${OUT}_metrics_summary.csv"
+
+# download counts h5 file
+wget https://cf.10xgenomics.com/samples/cell-exp/3.0.0/pbmc_1k_protein_v3/pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5 \
+  -O "${OUT}_filtered_feature_bc_matrix.h5"
+
+wget https://cf.10xgenomics.com/samples/cell-exp/3.0.0/pbmc_1k_protein_v3/pbmc_1k_protein_v3_raw_feature_bc_matrix.h5 \
+  -O "${OUT}_raw_feature_bc_matrix.h5"
+
+# download counts matrix tar gz file
+wget https://cf.10xgenomics.com/samples/cell-exp/3.0.0/pbmc_1k_protein_v3/pbmc_1k_protein_v3_filtered_feature_bc_matrix.tar.gz \
+  -O "${OUT}_filtered_feature_bc_matrix.tar.gz"
+
+# extract matrix tar gz
+mkdir -p "${OUT}_filtered_feature_bc_matrix"
+tar -xvf "${OUT}_filtered_feature_bc_matrix.tar.gz" \
+  -C "${OUT}_filtered_feature_bc_matrix" \
+  --strip-components 1
+rm "${OUT}_filtered_feature_bc_matrix.tar.gz"
+
+cat > /tmp/params.yaml << HERE
+  --input "${OUT}_filtered_feature_bc_matrix.h5" \
+  --input_metrics_summary "${OUT}_metrics_summary.csv" \
+  --output "${OUT}_filtered_feature_bc_matrix.h5mu"
+
+param_list:
+  - id: "$ID"
+    genome_fasta: "https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_41/GRCh38.primary_assembly.genome.fa.gz"
+    transcriptome_gtf: "https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_41/gencode.v41.annotation.gtf.gz"
+    target: ["bd_rhapsody", "cellranger_arc"] 
+    output_fasta: "reference.fa.gz"
+    output_gtf: "reference.gtf.gz"
+    non_nuclear_contigs: null
+    output_cellranger_arc: "reference_cellranger.tar.gz"
+    output_bd_rhapsody: "reference_bd_rhapsody.tar.gz"
+    bdrhap_extra_star_params: "--genomeSAindexNbases 12 --genomeSAsparseD 2"
+    motifs_file: "$motifs_modified"
+    subset_regex: "chr1"
+HERE
+
+# convert 10x h5 to h5mu
+nextflow run https://packages.viash-hub.com/vsh/openpipeline \
+  -latest \
+  -r v4.0.4 \
+  -main-script target/docker/convert/from_10xh5_to_h5mu/from_10xh5_to_h5mu \
+  -profile docker \
+  -c ./src/configs/labels_ci.config \
+  -params-file /tmp/params.yaml \
+  --publish_dir $OUT \
+  -resume
+
+# run single sample
+nextflow \
+  run . \
+  -main-script target/nextflow/workflows/rna/rna_singlesample/main.nf \
+  -c src/workflows/utils/labels_ci.config \
+  -profile docker \
+  --id pbmc_1k_protein_v3_uss \
+  --input "${OUT}_filtered_feature_bc_matrix.h5mu" \
+  --output "`basename $OUT`_uss.h5mu" \
+  --publishDir `dirname $OUT` \
+  -resume
+
+# add the sample ID to the mudata object
+nextflow \
+  run . \
+  -main-script target/nextflow/metadata/add_id/main.nf \
+  -c src/workflows/utils/labels_ci.config \
+  -profile docker \
+  --id pbmc_1k_protein_v3_uss \
+  --input "${OUT}_uss.h5mu" \
+  --input_id "pbmc_1k_protein_v3_uss" \
+  --output "`basename $OUT`_uss_with_id.h5mu" \
+  --output_compression "gzip" \
+  --publishDir `dirname $OUT` \
+  -resume
+
+# run multisample
+nextflow \
+  run . \
+  -main-script target/nextflow/workflows/rna/rna_multisample/main.nf \
+  -c src/workflows/utils/labels_ci.config \
+  -profile docker \
+  --id pbmc_1k_protein_v3_ums \
+  --input "${OUT}_uss_with_id.h5mu" \
+  --output "`basename $OUT`_ums.h5mu" \
+  --publishDir `dirname $OUT` \
+  -resume
+
+rm "${OUT}_uss_with_id.h5mu"
+
+# run dimred
+nextflow \
+  run . \
+  -main-script target/nextflow/workflows/multiomics/dimensionality_reduction/main.nf \
+  -c src/workflows/utils/labels_ci.config \
+  -profile docker \
+  --id pbmc_1k_protein_v3_mms \
+  --input "${OUT}_ums.h5mu" \
+  --output "`basename $OUT`_mms.h5mu" \
+  --publishDir `dirname $OUT` \
+  --obs_covariates sample_id \
+  -resume
+
+# run integration
+nextflow \
+  run . \
+  -main-script target/nextflow/workflows/integration/harmony_leiden/main.nf \
+  -c src/workflows/utils/labels_ci.config \
+  -profile docker \
+  --id pbmc_1k_protein_v3_mms_integration \
+  --input "${OUT}_mms.h5mu" \
+  --output "`basename $OUT`_mms.h5mu" \
+  --publishDir `dirname $OUT` \
+  --obs_covariates sample_id \
+  -resume
+
+python <<HEREDOC
+import mudata as mu
+mudata = mu.read_h5mu("${DIR}/pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu")
+mudata.mod["rna"].write_h5ad("${DIR}/pbmc_1k_protein_v3_filtered_feature_bc_matrix_rna.h5ad")
+HEREDOC
+
+aws s3 sync \
+  "$OUT" \
+  s3://openpipelines-bio/openpipeline_incubator/resources_test/"$ID" \
+  --exclude "*.yaml" \
+  --delete \
+  --dryrun
diff --git a/resources_test_scripts/qc_sample_data.sh b/resources_test_scripts/qc_sample_data.sh
new file mode 100755
index 0000000..ca97c4d
--- /dev/null
+++ b/resources_test_scripts/qc_sample_data.sh
@@ -0,0 +1,166 @@
+#/bin/bash
+
+OUT_DIR=resources_test/qc_sample_data
+OUT_DIR_SPATIAL=resources_test/spatial_qc_sample_data
+
+[ ! -d "$OUT_DIR" ] && mkdir -p "$OUT_DIR"
+[ ! -d "$OUT_DIR_SPATIAL" ] && mkdir -p "$OUT_DIR_SPATIAL"
+
+# fetch/create h5mu from somewhere
+cat > /tmp/params_create_h5mu.yaml <<EOF
+param_list:
+  - id: sample_one
+    input_id: sample_one
+    input: s3://openpipelines-data/10x_5k_anticmv/5k_human_antiCMV_T_TBNK_connect_qc.h5mu
+  - id: sample_two
+    input_id: sample_two
+    input: s3://openpipelines-data/10x_5k_anticmv/5k_human_antiCMV_T_TBNK_connect_qc.h5mu
+output: '\$id.qc.h5mu'
+output_compression: gzip
+publish_dir: "$OUT_DIR"
+EOF
+
+# add the sample ID to the mudata object
+nextflow run openpipelines-bio/openpipeline \
+  -latest \
+  -r 2.1.2 \
+  -main-script target/nextflow/metadata/add_id/main.nf \
+  -c src/configs/labels_ci.config \
+  -profile docker \
+  -params-file /tmp/params_create_h5mu.yaml \
+  -resume
+
+cat > /tmp/params_subset.yaml <<EOF
+param_list:
+  - id: sample_one
+    input: resources_test/qc_sample_data/sample_one.qc.h5mu
+  - id: sample_two
+    input: resources_test/qc_sample_data/sample_two.qc.h5mu
+output: '\$id.qc.h5mu'
+number_of_observations: 10000
+output_compression: gzip
+publish_dir: "$OUT_DIR"
+EOF
+
+# subset h5mus
+nextflow run openpipelines-bio/openpipeline \
+  -latest \
+  -r 2.1.2 \
+  -main-script target/nextflow/filter/subset_h5mu/main.nf \
+  -c src/configs/labels_ci.config \
+  -profile docker \
+  -params-file /tmp/params_subset.yaml \
+  -resume
+
+cat > /tmp/add_metadata_obs.py <<EOF
+import mudata as mu
+import glob
+import numpy as np
+import pandas as pd
+import os
+
+# Directory containing the h5mu files
+out_dir = "$(pwd)/resources_test/qc_sample_data"
+
+# List of h5mu files
+h5mu_files = glob.glob(os.path.join(out_dir, "*.h5mu"))
+print(f"Found {len(h5mu_files)} h5mu files: {h5mu_files}")
+
+# Metadata values to randomly assign
+donor_ids = ["donor_1", "donor_2", "donor_3"]
+cell_types = ["CD4+ T cell", "CD8+ T cell", "B cell", "NK cell", "Monocyte"]
+batches = ["batch_A", "batch_B"]
+conditions = ["treated", "control"]
+
+for h5mu_file in h5mu_files:
+    print(f"Processing {h5mu_file}...")
+    
+    # Load MuData object
+    mdata = mu.read_h5mu(h5mu_file)
+    rna = mdata.mod["rna"]
+    n_obs = rna.n_obs
+    
+    # Generate random metadata
+    np.random.seed(42 + hash(h5mu_file) % 100)  # Different seed for each file but reproducible
+    
+    # Create metadata
+    rna.obs["donor_id"] = np.random.choice(donor_ids, size=n_obs)
+    rna.obs["cell_type"] = np.random.choice(cell_types, size=n_obs)
+    rna.obs["batch"] = np.random.choice(batches, size=n_obs)
+    rna.obs["condition"] = np.random.choice(conditions, size=n_obs)
+    
+    # Add a continuous variable too
+    rna.obs["quality_score"] = np.random.uniform(0, 1, size=n_obs)
+    
+    # Save the modified MuData object
+    mu.write_h5mu(h5mu_file, mdata)
+    print(f"Added metadata to {h5mu_file}")
+
+print("All files processed successfully!")
+EOF
+
+# Execute the Python script
+python /tmp/add_metadata_obs.py
+
+# generate cellbender out for testing
+cat > /tmp/params_cellbender.yaml <<EOF
+param_list:
+  - id: sample_one
+    input: resources_test/qc_sample_data/sample_one.qc.h5mu
+  - id: sample_two
+    input: resources_test/qc_sample_data/sample_two.qc.h5mu
+output: '\$id.qc.cellbender.h5mu'
+epochs: 5
+output_compression: gzip
+publish_dir: "$OUT_DIR"
+EOF
+
+nextflow run openpipelines-bio/openpipeline \
+  -latest \
+  -r 2.1.2 \
+  -main-script target/nextflow/correction/cellbender_remove_background/main.nf \
+  -c src/configs/labels_ci.config \
+  -profile docker \
+  -params-file /tmp/params_cellbender.yaml \
+  -resume
+
+# fetch spatial sample data from s3
+aws s3 sync \
+  --profile di \
+  s3://openpipelines-bio/openpipeline_incubator/resources_test/spatial_qc_sample_data \
+  "$OUT_DIR_SPATIAL"
+
+# generate json for testing
+viash run src/ingestion_qc/h5mu_to_qc_json/config.vsh.yaml --engine docker -- \
+  --input "$OUT_DIR"/sample_one.qc.cellbender.h5mu \
+  --input "$OUT_DIR"/sample_two.qc.cellbender.h5mu \
+  --ingestion_method cellranger_multi \
+  --obs_metadata "donor_id;cell_type;batch;condition" \
+  --output "$OUT_DIR"/sc_dataset.json \
+  --output_reporting_json "$OUT_DIR"/sc_report_structure.json
+
+viash run src/ingestion_qc/h5mu_to_qc_json/config.vsh.yaml --engine docker -- \
+  --input "$OUT_DIR_SPATIAL"/xenium_tiny.qc.h5mu \
+  --input "$OUT_DIR_SPATIAL"/xenium_tiny.qc.h5mu \
+  --ingestion_method xenium \
+  --min_num_nonzero_vars 1 \
+  --output "$OUT_DIR_SPATIAL"/xenium_dataset.json \
+  --output_reporting_json "$OUT_DIR_SPATIAL"/xenium_report_structure.json
+
+# remove all state yaml files
+rm "$OUT_DIR"/*.yaml
+rm "$OUT_DIR_SPATIAL"/*.yaml
+
+# copy to s3
+aws s3 sync \
+  "$OUT_DIR" \
+  s3://openpipelines-bio/openpipeline_incubator/"$OUT_DIR" \
+  --delete \
+  --dryrun 
+
+
+aws s3 sync \
+  "$OUT_DIR_SPATIAL" \
+  s3://openpipelines-bio/openpipeline_incubator/"$OUT_DIR_SPATIAL" \
+  --delete \
+  --dryrun 
diff --git a/resources_test_scripts/reference_gencodev41.sh b/resources_test_scripts/reference_gencodev41.sh
new file mode 100644
index 0000000..39f9912
--- /dev/null
+++ b/resources_test_scripts/reference_gencodev41.sh
@@ -0,0 +1,74 @@
+#!/bin/bash
+
+set -eo pipefail
+
+# ensure that the command below is run from the root of the repository
+REPO_ROOT=$(git rev-parse --show-toplevel)
+cd "$REPO_ROOT"
+
+# settings
+ID=reference_gencodev41_chr1
+OUT=resources_test/$ID
+
+mkdir -p "$OUT" 
+
+wget "https://assets.thermofisher.com/TFS-Assets/LSG/manuals/ERCC92.zip" -O "$OUT/ERCC92.zip"
+
+# Download JASPAR files for reference building
+# Source of the code below: https://support.10xgenomics.com/single-cell-atac/software/release-notes/references#GRCh38-2020-A-2.0.0
+motifs_url="https://jaspar.elixir.no/download/data/2024/CORE/JASPAR2024_CORE_non-redundant_pfms_jaspar.txt"
+motifs_in="${OUT}/JASPAR2024_CORE_non-redundant_pfms_jaspar.txt"
+
+if [ ! -f "$motifs_in" ]; then
+    curl -sS "$motifs_url" > "$motifs_in"
+fi
+
+# Change motif headers so the human-readable motif name precedes the motif
+# identifier. So ">MA0004.1    Arnt" -> ">Arnt_MA0004.1".
+motifs_modified="${OUT}/$(basename "$motifs_in").modified"
+awk '{
+    if ( substr($1, 1, 1) == ">" ) {
+        print ">" $2 "_" substr($1,2)
+    } else {
+        print
+    }
+}' "$motifs_in" > "$motifs_modified"
+
+
+cat > /tmp/params.yaml << HERE
+param_list:
+  - id: "$ID"
+    genome_fasta: "https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_41/GRCh38.primary_assembly.genome.fa.gz"
+    transcriptome_gtf: "https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_41/gencode.v41.annotation.gtf.gz"
+    target: ["bd_rhapsody", "cellranger_arc"] 
+    output_fasta: "reference.fa.gz"
+    output_gtf: "reference.gtf.gz"
+    non_nuclear_contigs: null
+    output_cellranger_arc: "reference_cellranger.tar.gz"
+    output_bd_rhapsody: "reference_bd_rhapsody.tar.gz"
+    bdrhap_extra_star_params: "--genomeSAindexNbases 12 --genomeSAsparseD 2"
+    motifs_file: "$motifs_modified"
+    subset_regex: "chr1"
+HERE
+
+nextflow run https://packages.viash-hub.com/vsh/openpipeline \
+  -latest \
+  -r v4.0.4 \
+  -main-script target/nextflow/workflows/ingestion/make_reference/main.nf \
+  -profile docker \
+  -c ./src/configs/labels_ci.config \
+  -params-file /tmp/params.yaml \
+  --publish_dir $OUT \
+  -resume
+
+rm "$motifs_modified"
+rm "$motifs_in"
+rm "$OUT/ERCC92.zip"
+
+
+aws s3 sync \
+  "$OUT" \
+  s3://openpipelines-bio/openpipeline_incubator/resources_test/"$ID" \
+  --exclude "*.yaml" \
+  --delete \
+  --dryrun
diff --git a/resources_test_scripts/spatial_qc_sample_data.sh b/resources_test_scripts/spatial_qc_sample_data.sh
new file mode 100755
index 0000000..dca390e
--- /dev/null
+++ b/resources_test_scripts/spatial_qc_sample_data.sh
@@ -0,0 +1,37 @@
+#/bin/bash
+
+OUT_DIR=resources_test/spatial_qc_sample_data
+
+[ ! -d "$OUT_DIR" ] && mkdir -p "$OUT_DIR"
+
+# fetch/create h5mu from somewhere
+cat > /tmp/qc.yaml <<EOF
+param_list:
+  - id: xenium_tiny
+    input: s3://openpipelines-bio/openpipeline_spatial/resources_test/xenium/xenium_tiny.h5mu
+  - id: Lung5_Rep2_tiny
+    input: s3://openpipelines-bio/openpipeline_spatial/resources_test/cosmx/Lung5_Rep2_tiny.h5mu
+var_name_mitochondrial_genes: mitochondrial
+var_name_ribosomal_genes: ribosomal
+output: '\$id.qc.h5mu'
+output_compression: gzip
+publish_dir: "$OUT_DIR"
+EOF
+
+nextflow run openpipelines-bio/openpipeline \
+  -latest \
+  -r 2.1.0 \
+  -main-script target/nextflow/workflows/qc/qc/main.nf \
+  -profile docker \
+  -params-file /tmp/qc.yaml \
+  -resume \
+  -config src/configs/labels_ci.config
+
+# copy to s3
+aws s3 sync \
+  --profile di \
+  resources_test/spatial_qc_sample_data \
+  s3://openpipelines-bio/openpipeline_incubator/resources_test/spatial_qc_sample_data \
+  --delete --dryrun \
+  --exclude "*" --include "*.h5mu" \
+  
diff --git a/src/authors/dorien_roosen.yaml b/src/authors/dorien_roosen.yaml
new file mode 100644
index 0000000..b23fca0
--- /dev/null
+++ b/src/authors/dorien_roosen.yaml
@@ -0,0 +1,11 @@
+name: Dorien Roosen
+info:
+  role: Core Team Member
+  links:
+    email: dorien@data-intuitive.com
+    github: dorien-er
+    linkedin: dorien-roosen
+  organizations:
+    - name: Data Intuitive
+      href: https://www.data-intuitive.com
+      role: Data Scientist
diff --git a/src/authors/jakub_majercik.yaml b/src/authors/jakub_majercik.yaml
new file mode 100644
index 0000000..1dac285
--- /dev/null
+++ b/src/authors/jakub_majercik.yaml
@@ -0,0 +1,11 @@
+name: Jakub Majercik
+info:
+  role: Contributor
+  links:
+    email: jakub@data-intuitive.com
+    github: jakubmajercik
+    linkedin: jakubmajercik
+  organizations:
+    - name: Data Intuitive
+      href: https://www.data-intuitive.com
+      role: Bioinformatics Engineer
diff --git a/src/authors/robrecht_cannoodt.yaml b/src/authors/robrecht_cannoodt.yaml
new file mode 100644
index 0000000..8f0fe78
--- /dev/null
+++ b/src/authors/robrecht_cannoodt.yaml
@@ -0,0 +1,15 @@
+name: Robrecht Cannoodt
+info:
+  role: Core Team Member
+  links:
+    email: robrecht@data-intuitive.com
+    github: rcannood
+    orcid: "0000-0003-3641-729X"
+    linkedin: robrechtcannoodt
+  organizations:
+    - name: Data Intuitive
+      href: https://www.data-intuitive.com
+      role: Data Science Engineer
+    - name: Open Problems
+      href: https://openproblems.bio
+      role: Core Member
diff --git a/src/authors/weiwei_schultz.yaml b/src/authors/weiwei_schultz.yaml
new file mode 100644
index 0000000..04d8904
--- /dev/null
+++ b/src/authors/weiwei_schultz.yaml
@@ -0,0 +1,6 @@
+name: Weiwei Schultz
+info:
+  role: Contributor
+  organizations:
+    - name: Janssen R&D US
+      role: Associate Director Data Sciences
diff --git a/src/configs/integration_tests.config b/src/configs/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/src/configs/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/src/configs/labels.config b/src/configs/labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/src/configs/labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/src/configs/labels_ci.config b/src/configs/labels_ci.config
new file mode 100644
index 0000000..7a1c7d3
--- /dev/null
+++ b/src/configs/labels_ci.config
@@ -0,0 +1,105 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
+
+profiles {
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.fixOwnership    = true
+    docker.enabled         = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+
+  local {
+    // This config is for local processing.
+    process {
+        maxMemory = 25.GB
+        withLabel: verylowcpu { cpus = 2 }
+        withLabel: lowcpu { cpus = 4 }
+        withLabel: midcpu { cpus = 6 }
+        withLabel: highcpu { cpus = 12 }
+  
+        withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+        withLabel: midmem { memory = { get_memory( 12.GB * task.attempt ) } }
+        withLabel: highmem { memory = { get_memory( 20.GB * task.attempt ) } }
+    }
+  }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/src/single_cell/cellranger_multi_qc/cellranger_multi.yaml b/src/single_cell/cellranger_multi_qc/cellranger_multi.yaml
new file mode 100644
index 0000000..bf2d1dd
--- /dev/null
+++ b/src/single_cell/cellranger_multi_qc/cellranger_multi.yaml
@@ -0,0 +1,372 @@
+argument_groups:
+  - name: Input files
+    arguments:
+      - type: file
+        name: --input
+        required: false
+        description: |
+          The FASTQ files to be analyzed. FASTQ files should conform to the naming conventions of bcl2fastq and mkfastq:
+          `[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`
+        example: [ mysample_S1_L001_R1_001.fastq.gz, mysample_S1_L001_R2_001.fastq.gz ]
+        multiple: true
+
+  - name: Library arguments
+    arguments:
+      - type: string
+        name: --library_id
+        required: false
+        description: |
+          The Illumina sample name to analyze. This must exactly match the 'Sample Name'part
+          of the FASTQ files specified in the `--input` argument.
+        example: ["mysample1"]
+        multiple: true
+      - type: string
+        name: --library_type
+        required: false
+        description: |
+          The underlying feature type of the library.
+        choices: ["Gene Expression", "VDJ", "VDJ-T", "VDJ-B", "VDJ-T-GD", "Antibody Capture",
+                    "CRISPR Guide Capture", "Multiplexing Capture", "Antigen Capture", "Custom"]
+        example: "Gene Expression"
+        multiple: true
+      - type: string
+        name: --library_subsample
+        required: false
+        description: |
+          The rate at which reads from the provided FASTQ files are sampled.
+          Must be strictly greater than 0 and less than or equal to 1.
+        example: "0.5"
+        multiple: true
+      - type: string
+        name: --library_lanes
+        required: false
+        description: Lanes associated with this sample. Defaults to using all lanes.
+        example: "1-4"
+        multiple: true
+      - type: string
+        name: "--library_chemistry"
+        description: |
+          Only applicable to FRP. Library-specific assay configuration. By default,
+          the assay configuration is detected automatically. Typically, users will
+          not need to specify a chemistry.
+
+  - name: Sample parameters
+    # Corresponds to the [samples] section
+    arguments:
+      - type: string
+        name: --sample_ids
+        alternatives: "--cell_multiplex_sample_id"
+        multiple: true
+        description: |
+          A name to identify a multiplexed sample. Must be alphanumeric with hyphens and/or underscores,
+          and less than 64 characters. Required for Cell Multiplexing libraries.
+      - type: string
+        multiple: true
+        name: --sample_description
+        alternatives: [--cell_multiplex_description]
+        description: A description for the sample.
+      - type: integer
+        multiple: true
+        name: --sample_expect_cells
+        example: 3000
+        description: |
+          Expected number of recovered cells, used as input to cell calling algorithm.
+      - type: integer
+        name: "--sample_force_cells"
+        example: 3000
+        multiple: true
+        required: false
+        description: |
+          Force pipeline to use this number of cells, bypassing cell detection.
+
+  - name: "Feature Barcode library specific arguments"
+    # Corresponds to the [feature] section
+    arguments:
+      - name: "--feature_reference"
+        type: file
+        description: |
+          Path to the Feature reference CSV file, declaring Feature Barcode constructs and associated barcodes.
+          Required only for Antibody Capture or CRISPR Guide Capture libraries.
+          See https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/using/feature-bc-analysis#feature-ref for more information."
+        example: "feature_reference.csv"
+        required: false
+      - name: "--feature_r1_length"
+        type: integer
+        required: false
+        description: |
+          Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,
+          where N is the user-supplied value. Note that the length includes the Barcode and UMI
+          sequences so do not set this below 26.
+      - name: "--feature_r2_length"
+        type: integer
+        required: false
+        description: |
+          Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,
+          where N is a user-supplied value. Trimming occurs before sequencing metrics are computed
+          and therefore, limiting the length of Read 2 may affect Q30 scores.
+      - name: "--min_crispr_umi"
+        type: integer
+        min: 1
+        required: false
+        description: |
+          Set the minimum number of CRISPR guide RNA UMIs required for protospacer detection.
+          If a lower or higher sensitivity is desired for detection, this value can be customized
+          according to specific experimental needs. Applicable only to datasets that include a
+          CRISPR Guide Capture library.
+  - name: Gene expression arguments
+    # Corresponds to the [gene-expression] section
+    description: Arguments relevant to the analysis of gene expression data.
+    arguments:
+      - name: "--gex_reference"
+        type: file
+        description: "Genome refence index built by Cell Ranger mkref."
+        example: "reference_genome.tar.gz"
+        required: true
+      - type: boolean
+        name: "--gex_secondary_analysis"
+        default: false
+        description: Whether or not to run the secondary analysis e.g. clustering.
+      - type: boolean
+        name: "--gex_generate_bam"
+        default: false
+        description: Whether to generate a BAM file.
+      - type: file
+        name: "--tenx_cloud_token_path"
+        description: The 10x Cloud Analysis user token used to enable cell annotation.
+      - type: string
+        name: "--cell_annotation_model"
+        description: |
+          "Cell annotation model to use. If auto, uses the default model for the species.
+          If not given, does not run cell annotation."
+        choices: ["auto", "human_pca_v1_beta", "mouse_pca_v1_beta"]
+      - type: integer
+        name: --gex_expect_cells
+        example: 3000
+        description: |
+          Expected number of recovered cells, used as input to cell calling algorithm.
+      - type: integer
+        name: "--gex_force_cells"
+        example: 3000
+        description: |
+          Force pipeline to use this number of cells, bypassing cell detection.
+      - type: boolean
+        name: "--gex_include_introns"
+        default: true
+        description: |
+          Whether or not to include intronic reads in counts.
+          This option does not apply to Fixed RNA Profiling analysis.
+      - name: "--gex_r1_length"
+        type: integer
+        required: false
+        description: |
+          Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,
+          where N is the user-supplied value. Note that the length includes the Barcode and UMI
+          sequences so do not set this below 26.
+      - name: "--gex_r2_length"
+        type: integer
+        required: false
+        description: |
+          Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,
+          where N is a user-supplied value. Trimming occurs before sequencing metrics are computed
+          and therefore, limiting the length of Read 2 may affect Q30 scores.
+      - type: string
+        name: --gex_chemistry
+        default: auto
+        description: |
+          Assay configuration. Either specify a single value which will be applied to all libraries,
+          or a number of values that is equal to the number of libararies. The latter is only applicable
+          to only applicable to Fixed RNA Profiling.
+            - auto: Chemistry autodetection (default)
+            - threeprime: Single Cell 3'
+            - SC3Pv1, SC3Pv2, SC3Pv3(-polyA), SC3Pv4(-polyA): Single Cell 3' v1, v2, v3, or v4
+            - SC3Pv3HT(-polyA): Single Cell 3' v3.1 HT
+            - SC-FB: Single Cell Antibody-only 3' v2 or 5'
+            - fiveprime: Single Cell 5'
+            - SC5P-PE: Paired-end Single Cell 5'
+            - SC5P-PE-v3: Paired-end Single Cell 5' v3
+            - SC5P-R2: R2-only Single Cell 5'
+            - SC5P-R2-v3: R2-only Single Cell 5' v3
+            - SCP5-PE-v3: Single Cell 5' paired-end v3 (GEM-X)
+            - SC5PHT : Single Cell 5' v2 HT
+            - SFRP: Fixed RNA Profiling (Singleplex)
+            - MFRP: Fixed RNA Profiling (Multiplex, Probe Barcode on R2)
+            - MFRP-R1: Fixed RNA Profiling (Multiplex, Probe Barcode on R1)
+            - MFRP-RNA: Fixed RNA Profiling (Multiplex, RNA, Probe Barcode on R2)
+            - MFRP-Ab: Fixed RNA Profiling (Multiplex, Antibody, Probe Barcode at R2:69)
+            - MFRP-Ab-R2pos50: Fixed RNA Profiling (Multiplex, Antibody, Probe Barcode at R2:50)
+            - MFRP-RNA-R1: Fixed RNA Profiling (Multiplex, RNA, Probe Barcode on R1)
+            - MFRP-Ab-R1: Fixed RNA Profiling (Multiplex, Antibody, Probe Barcode on R1)
+            - ARC-v1 for analyzing the Gene Expression portion of Multiome data. If Cell Ranger auto-detects ARC-v1 chemistry, an error is triggered.
+          See https://kb.10xgenomics.com/hc/en-us/articles/115003764132-How-does-Cell-Ranger-auto-detect-chemistry- for more information.
+        choices: [ auto, threeprime, fiveprime, SC3Pv1, SC3Pv2, SC3Pv3, SC3Pv3-polyA, SC3Pv4, SC3Pv4-polyA, SC3Pv3LT, SC3Pv3HT, SC3Pv3HT-polyA, 
+                  SC5P-PE, SC5P-PE-v3, SC5P-R2, SC-FB, SC5P-R2-v3, SCP5-PE-v3, SC5PHT, MFRP, MFRP-R1, MFRP-RNA, MFRP-Ab,
+                  SFRP, MFRP-Ab-R2pos50, MFRP-RNA-R1, MFRP-Ab-R1, ARC-v1]
+
+  - name: "VDJ related parameters"
+    # The [vdj] section
+    arguments:
+      - name: "--vdj_reference"
+        type: file
+        description: "VDJ refence index built by Cell Ranger mkref."
+        example: "reference_vdj.tar.gz"
+        required: false
+      - name: "--vdj_inner_enrichment_primers"
+        type: file
+        description: |
+          V(D)J Immune Profiling libraries: if inner enrichment primers other than those provided 
+          in the 10x Genomics kits are used, they need to be specified here as a
+          text file with one primer per line.
+        example: "enrichment_primers.txt"
+        required: false
+      - name: "--vdj_r1_length"
+        type: integer
+        required: false
+        description: |
+          Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases, where N is the user-supplied value.
+          Note that the length includes the Barcode and UMI sequences so do not set this below 26.
+      - name: "--vdj_r2_length"
+        type: integer
+        required: false
+        description: |
+          Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases, where N is a user-supplied value. 
+          Trimming occurs before sequencing metrics are computed and therefore, limiting the length of Read 2 may affect Q30 scores
+      - name: "--vdj_denovo"
+        type: boolean
+        required: false
+        description: |
+          Run in reference-free mode (i.e., do not use annotations). This option is not supported for multiplexed experiments.
+  - name: 3' Cell multiplexing parameters (CellPlex Multiplexing)
+    # cell_multiplex_oligo_ids adds to [samples] section
+    # min_assignment_confidence, cmo_set barcode_sample_assignment are added to [gene-expression]
+    arguments:
+      - type: string
+        name: --cell_multiplex_oligo_ids
+        alternatives: [--cmo_ids]
+        multiple: true
+        description: |
+          The Cell Multiplexing oligo IDs used to multiplex this sample. If multiple CMOs were used for a sample,
+          separate IDs with a pipe (e.g., CMO301|CMO302). Required for Cell Multiplexing libraries.
+        
+      - type: double
+        name: --min_assignment_confidence
+        description: | 
+          The minimum estimated likelihood to call a sample as tagged with a Cell Multiplexing Oligo (CMO) instead of "Unassigned".
+          Users may wish to tolerate a higher rate of mis-assignment in order to obtain more singlets to include in their analysis,
+          or a lower rate of mis-assignment at the cost of obtaining fewer singlets.
+      - type: file 
+        direction: input
+        required: false
+        name: "--cmo_set"
+        description: |
+          Path to a custom CMO set CSV file, declaring CMO constructs and associated barcodes. If the default CMO reference IDs that are built into
+          the Cell Ranger software are required, this option does not need to be used.
+      - type: file
+        direction: input
+        required: false
+        name: "--barcode_sample_assignment"
+        description: |
+          Path to a barcode-sample assignment CSV file that specifies the barcodes that belong to each sample.
+  
+  - name: Hashtag multiplexing parameters
+    # Is added to [samples]
+    arguments:
+      - name: --hashtag_ids
+        type: string
+        multiple: true
+        description: |
+          The hashtag IDs used to multiplex this sample. If multiple antibody hashtags were used for the same sample,
+          you can separate IDs with a pipe.
+
+  - name: On-chip multiplexing parameters
+    # Is added to [samples]
+    arguments:
+      - name: --ocm_barcode_ids
+        type: string
+        multiple: true
+        # Note: choices is not an option here because multiple values can be added using pipe
+        description: |
+          The OCM barcode IDs used to multiplex this sample. Must be one of OB1, OB2, OB3, OB4.
+          If multiple OCM Barcodes were used for the same sample, you can separate IDs
+          with a pipe (e.g., OB1|OB2).
+
+  - name: Flex multiplexing paramaters
+    # probe_set, filter_probes and emptydrops_minimum_umis end up in [gene-expression]
+    # probe_barcode_ids ends up in [samples]
+    arguments:
+      - type: file
+        name: "--probe_set"
+        description: |
+          A probe set reference CSV file. It specifies the sequences used as a reference for probe alignment and the gene ID associated with each probe.
+          It must include 4 columns (probe file format 1.0.0): gene_id,probe_seq,probe_id,included,region and an optional 5th column (probe file format 1.0.1).
+          - gene_id: The Ensembl gene identifier targeted by the probe.
+          - probe_seq: The nucleotide sequence of the probe, which is complementary to the transcript sequence.
+          - probe_id: The probe identifier, whose format is described in Probe identifiers.
+          - included: A TRUE or FALSE flag specifying whether the probe is included in the filtered counts matrix output or excluded by the probe filter. 
+                      See filter-probes option of cellranger multi. All probes of a gene must be marked TRUE in the included column for that gene to be included.
+          - region: Present only in v1.0.1 probe set reference CSV. The gene boundary targeted by the probe. Accepted values are spliced or unspliced.
+
+          The file also contains a number of required metadata fields in the header in the format #key=value:
+          - panel_name: The name of the probe set.
+          - panel_type: Always predesigned for predesigned probe sets.
+          - reference_genome: The reference genome build used for probe design.
+          - reference_version: The version of the Cell Ranger reference transcriptome used for probe design.
+          - probe_set_file_format: The version of the probe set file format specification that this file conforms to.
+      - type: boolean  # Null is also a valid option because passing this argument to cellranger (true or false) requires --probe_set 
+        name: "--filter_probes"
+        description: |
+          If 'false', include all non-deprecated probes listed in the probe set reference CSV file.
+          If 'true' or not set, probes that are predicted to have off-target activity to homologous genes are excluded from analysis.
+          Not filtering will result in UMI counts from all non-deprecated probes,
+          including those with predicted off-target activity, to be used in the analysis.
+          Probes whose ID is prefixed with DEPRECATED are always excluded from the analysis.
+
+      - type: string
+        name: "--probe_barcode_ids"
+        multiple: true
+        description: |
+          The Fixed RNA Probe Barcode ID used for this sample, and for multiplex GEX + Antibody Capture libraries,
+          the corresponding Antibody Multiplexing Barcode IDs. 10x recommends specifying both barcodes (e.g., BC001+AB001)
+          when an Antibody Capture library is present. The barcode pair order is BC+AB and they
+          are separated with a "+" (no spaces). Alternatively, you can specify the Probe Barcode ID alone and
+          Cell Ranger's barcode pairing auto-detection algorithm will automatically match to the corresponding Antibody
+          Multiplexing Barcode.
+      - type: integer
+        name: --emptydrops_minimum_umis
+        min: 1
+        description: |
+          For singleplex Flex experiments, use this option to adjust the UMI cutoff during the second step of cell calling.
+          Cell Ranger will still perform the full cell calling process but will only evaluate barcodes with UMIs above
+          the threshold you specify.
+
+  - name: Antigen Capture (BEAM) libary arguments
+    # These end up in the [antigen-specificity] section
+    description: |
+      These arguments are recommended if an Antigen Capture (BEAM) library is present. 
+      It is needed to calculate the antigen specificity score.
+    arguments:
+      - type: string
+        name: --control_id
+        multiple: true
+        description: |
+          A user-defined ID for any negative controls used in the T/BCR Antigen Capture assay. Must match id specified in the feature reference CSV.
+          May only include ASCII characters and must not use whitespace, slash, quote, or comma characters. 
+          Each ID must be unique and must not collide with a gene identifier from the transcriptome.
+      - type: string
+        multiple: true
+        name: --mhc_allele
+        description: |
+          The MHC allele for TCR Antigen Capture libraries. Must match mhc_allele name specified in the Feature Reference CSV.
+  - name: "General arguments"
+    description: |
+      These arguments are applicable to all library types.
+    arguments:
+      - name: "--check_library_compatibility"
+        type: boolean
+        default: true
+        description: |
+          Optional. This option allows users to disable the check that evaluates 10x Barcode overlap between
+          ibraries when multiple libraries are specified (e.g., Gene Expression + Antibody Capture). Setting
+          this option to false will disable the check across all library combinations. We recommend running
+          this check (default), however if the pipeline errors out, users can bypass the check to generate
+          outputs for troubleshooting.
+
diff --git a/src/single_cell/cellranger_multi_qc/config.vsh.yaml b/src/single_cell/cellranger_multi_qc/config.vsh.yaml
new file mode 100644
index 0000000..8595ffb
--- /dev/null
+++ b/src/single_cell/cellranger_multi_qc/config.vsh.yaml
@@ -0,0 +1,101 @@
+name: "cellranger_multi_qc"
+namespace: "single_cell"
+# scope: "private"
+description: "A pipeline for running Cell Ranger multi followed by QC."
+authors:
+  - __merge__: /src/authors/jakub_majercik.yaml
+    roles: [ author, maintainer ]
+  - __merge__: /src/authors/weiwei_schultz.yaml
+    roles: [ contributor ]
+__merge__: /src/single_cell/cellranger_multi_qc/cellranger_multi.yaml
+argument_groups:
+  - name: Outputs
+    arguments:
+      - name: "--output_raw"
+        type: file
+        direction: output
+        description: "The raw output folder."
+        required: true
+        example: output_dir/
+      - name: "--output_h5mu"
+        type: file
+        direction: output
+        description: |
+          Locations for the output files. Must contain a wildcard (*) character,
+          which will be replaced with the sample name.
+        example: "*.h5mu"
+        required: true
+      - name: "--uns_metrics"
+        type: string
+        description: Name of the .uns slot under which to QC metrics (if any).
+        default: "metrics_cellranger"
+      - name: "--output_ingestion_qc_report"
+        type: file
+        direction: output
+        required: false
+        multiple: true
+        description: |
+          Ingestion QC report in HTML format. Generated when --create_sample_qc_report is true.
+        example: "*.sample_qc_report.html"
+      - name: "--output_processed_h5mu"
+        type: file
+        direction: output
+        required: false
+        description: |
+          Folder containing the QC-processed h5mu files. Generated when
+          --create_sample_qc_report is true.
+        example: "processed_h5mu/"
+      - name: "--output_multiqc_report"
+        type: file
+        direction: output
+        required: false
+        description: |
+          MultiQC report in HTML format. Generated when
+          --create_multiqc_report is true.
+        example: "multiqc_report.html"
+
+  - name: "QC reports"
+    description: "Options for generating optional QC reports."
+    arguments:
+      - name: "--create_sample_qc_report"
+        type: boolean
+        default: true
+        description: |
+          Whether to generate an ingestion QC report.
+      - name: "--create_multiqc_report"
+        type: boolean
+        default: true
+        description: |
+          Whether to run FastQC on the input FASTQ files and aggregate results
+          into a MultiQC report.
+      - name: "--run_cellbender"
+        type: boolean
+        default: false
+        description: |
+          Whether to run CellBender for ambient RNA removal as part of the
+          sample QC report generation. Only used when --create_sample_qc_report is true.
+
+dependencies:
+  - name: workflows/ingestion/cellranger_multi
+    repository: openpipeline
+  - name: workflows/generate_qc_report
+    repository: openpipeline_qc
+  - name: fastqc
+    repository: biobox
+  - name: multiqc
+    repository: biobox
+
+resources:
+  - type: nextflow_script
+    path: main.nf
+    entrypoint: run_wf
+
+test_resources:
+  - type: nextflow_script
+    path: test.nf
+    entrypoint: test_wf
+  - path: /resources_test/10x_5k_anticmv/raw/
+  - path: /resources_test/10x_5k_fixed/raw/
+  - path: /resources_test/reference_gencodev41_chr1
+runners:
+  - type: nextflow
diff --git a/src/single_cell/cellranger_multi_qc/integration_test.sh b/src/single_cell/cellranger_multi_qc/integration_test.sh
new file mode 100755
index 0000000..f6e05b4
--- /dev/null
+++ b/src/single_cell/cellranger_multi_qc/integration_test.sh
@@ -0,0 +1,38 @@
+#!/bin/bash
+
+set -eo pipefail
+
+# get the root of the directory
+REPO_ROOT=$(git rev-parse --show-toplevel)
+
+# ensure that the command below is run from the root of the repository
+cd "$REPO_ROOT"
+
+COMMON_ARGS=(
+  run .
+  -main-script src/single_cell/cellranger_multi_qc/test.nf
+  -resume
+  -profile docker
+  -c src/configs/labels_ci.config
+  -c src/configs/integration_tests.config
+)
+
+# test_wf: GEX + AB, sample QC report only
+nextflow "${COMMON_ARGS[@]}" \
+  -entry test_wf \
+  --publish_dir test_output/cellranger_multi_qc/test_wf
+
+# test_wf_ab_only: AB-only input, all report steps skipped
+nextflow "${COMMON_ARGS[@]}" \
+  -entry test_wf_ab_only \
+  --publish_dir test_output/cellranger_multi_qc/test_wf_ab_only
+
+# test_wf_both_reports: GEX + AB, both MultiQC and sample QC reports
+nextflow "${COMMON_ARGS[@]}" \
+  -entry test_wf_both_reports \
+  --publish_dir test_output/cellranger_multi_qc/test_wf_both_reports
+
+# test_wf_multiqc_only: GEX + AB, MultiQC report only
+nextflow "${COMMON_ARGS[@]}" \
+  -entry test_wf_multiqc_only \
+  --publish_dir test_output/cellranger_multi_qc/test_wf_multiqc_only
diff --git a/src/single_cell/cellranger_multi_qc/main.nf b/src/single_cell/cellranger_multi_qc/main.nf
new file mode 100644
index 0000000..447eaba
--- /dev/null
+++ b/src/single_cell/cellranger_multi_qc/main.nf
@@ -0,0 +1,81 @@
+workflow run_wf {
+  take:
+    input_ch
+
+  main:
+    output_ch = input_ch
+      | map { id, state ->
+          [id, state + [_meta: [join_id: id]]]
+        }
+
+      | fastqc.run(
+          runIf: { id, state -> state.create_multiqc_report && state.library_type?.contains("Gene Expression") },
+          fromState: { id, state ->
+            [
+              input: state.input,
+              outdir: "${id}_fastqc"
+            ]
+          },
+          toState: { id, output, state ->
+            state + [output_fastqc: output.outdir]
+          }
+        )
+
+      | cellranger_multi.run(
+          fromState: { id, state -> state },
+          toState: { id, output, state ->
+            state + [
+              output_raw: output.output_raw,
+              output_h5mu: output.output_h5mu
+            ]
+          }
+        )
+
+      | multiqc.run(
+          runIf: { id, state -> state.create_multiqc_report && state.library_type?.contains("Gene Expression") },
+          fromState: { id, state ->
+            [
+              input: [state.output_fastqc, state.output_raw],
+              output_report: state.output_multiqc_report
+            ]
+          },
+          toState: { id, output, state ->
+            state + [_multiqc_produced: true, output_multiqc_report: output.output_report]
+          }
+        )
+      
+      | generate_qc_report.run(
+          runIf: { id, state -> state.create_sample_qc_report && state.library_type?.contains("Gene Expression") },
+          fromState: { id, state ->
+            [
+              id: id,
+              input: state.output_h5mu,
+              ingestion_method: "cellranger_multi",
+              run_cellbender: state.run_cellbender,
+              output_qc_report: state.output_ingestion_qc_report,
+              output_processed_h5mu: state.output_processed_h5mu
+            ]
+          },
+          toState: { id, output, state ->
+            state + [
+              _qc_report_produced: true,
+              output_ingestion_qc_report: output.output_qc_report,
+              output_processed_h5mu: output.output_processed_h5mu
+            ]
+          }
+        )
+
+      | map { id, state ->
+          def out = [output_raw: state.output_raw, output_h5mu: state.output_h5mu]
+          if (state._meta) out._meta = state._meta
+          if (state._multiqc_produced) out.output_multiqc_report = state.output_multiqc_report
+          if (state._qc_report_produced) {
+            out.output_ingestion_qc_report = state.output_ingestion_qc_report
+            out.output_processed_h5mu = state.output_processed_h5mu
+          }
+          [id, out]
+        }
+
+  emit:
+    output_ch
+}
diff --git a/src/single_cell/cellranger_multi_qc/nextflow.config b/src/single_cell/cellranger_multi_qc/nextflow.config
new file mode 100644
index 0000000..21e68ab
--- /dev/null
+++ b/src/single_cell/cellranger_multi_qc/nextflow.config
@@ -0,0 +1,10 @@
+manifest {
+  nextflowVersion = '!>=20.12.1-edge'
+}
+
+params {
+  rootDir = java.nio.file.Paths.get("$projectDir/../../../").toAbsolutePath().normalize().toString()
+}
+
+// include common settings
+includeConfig("${params.rootDir}/src/configs/labels.config")
diff --git a/src/single_cell/cellranger_multi_qc/test.nf b/src/single_cell/cellranger_multi_qc/test.nf
new file mode 100644
index 0000000..f3fbf51
--- /dev/null
+++ b/src/single_cell/cellranger_multi_qc/test.nf
@@ -0,0 +1,210 @@
+nextflow.enable.dsl=2
+
+include { cellranger_multi_qc } from params.rootDir + "/target/nextflow/single_cell/cellranger_multi_qc/main.nf"
+
+params.resources_test = "s3://openpipelines-bio/openpipeline_incubator/resources_test/"
+
+workflow test_wf {
+  resources_test = file(params.resources_test)
+
+  output_ch = Channel.fromList([
+      [
+        id: "sample_anticmv",
+        input: [
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_GEX_1_subset_S1_L001_R1_001.fastq.gz"),
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_GEX_1_subset_S1_L001_R2_001.fastq.gz"),
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_AB_subset_S2_L004_R1_001.fastq.gz"),
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_AB_subset_S2_L004_R2_001.fastq.gz")
+        ],
+        gex_reference: resources_test.resolve("reference_gencodev41_chr1/reference_cellranger.tar.gz"),
+        feature_reference: resources_test.resolve("10x_5k_anticmv/raw/feature_reference.csv"),
+        library_id: ["5k_human_antiCMV_T_TBNK_connect_GEX_1_subset", "5k_human_antiCMV_T_TBNK_connect_AB_subset"],
+        library_type: ["Gene Expression", "Antibody Capture"],
+        output_raw: "sample_anticmv_raw/",
+        output_h5mu: "sample_anticmv.h5mu",
+        create_sample_qc_report: true,
+        output_ingestion_qc_report: "sample_anticmv_qc_report_*.html",
+        output_processed_h5mu: "sample_anticmv_processed"
+      ]
+    ])
+    | map { state -> [state.id, state] }
+    | cellranger_multi_qc
+    | view { output ->
+      assert output.size() == 2 : "Outputs should contain two elements; [id, state]"
+
+      def id = output[0]
+      assert id == "combined" : "Output ID should be 'combined'. Found: ${id}"
+
+      def state = output[1]
+      assert state instanceof Map : "State should be a map. Found: ${state}"
+
+      assert state.containsKey("output_raw") : "State should contain key 'output_raw'."
+      assert state.output_raw.isDirectory() : "'output_raw' should be a directory."
+
+      assert state.containsKey("output_h5mu") : "State should contain key 'output_h5mu'."
+      assert state.output_h5mu.isFile() : "'output_h5mu' should be a file."
+      assert state.output_h5mu.toString().endsWith(".h5mu") : "output_h5mu should end with '.h5mu'. Found: ${state.output_h5mu}"
+
+      assert state.containsKey("output_ingestion_qc_report") : "State should contain key 'output_ingestion_qc_report'."
+      assert state.output_ingestion_qc_report instanceof List : "'output_ingestion_qc_report' should be a list."
+      assert state.output_ingestion_qc_report.every { it.isFile() } : "All QC report files should exist."
+
+      assert state.containsKey("output_processed_h5mu") : "State should contain key 'output_processed_h5mu'."
+      assert state.output_processed_h5mu.isDirectory() : "'output_processed_h5mu' should be a directory."
+
+      "Output: $output"
+    }
+}
+
+workflow test_wf_ab_only {
+  resources_test = file(params.resources_test)
+
+  output_ch = Channel.fromList([
+      [
+        id: "sample_ab_only",
+        input: [
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_AB_subset_S2_L004_R1_001.fastq.gz"),
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_AB_subset_S2_L004_R2_001.fastq.gz")
+        ],
+        gex_reference: resources_test.resolve("reference_gencodev41_chr1/reference_cellranger.tar.gz"),
+        feature_reference: resources_test.resolve("10x_5k_anticmv/raw/feature_reference.csv"),
+        library_id: ["5k_human_antiCMV_T_TBNK_connect_AB_subset"],
+        library_type: ["Antibody Capture"],
+        output_raw: "sample_ab_only_raw/",
+        output_h5mu: "sample_ab_only.h5mu",
+        create_sample_qc_report: true,
+        create_multiqc_report: true,
+        output_ingestion_qc_report: "sample_ab_only_qc_report_*.html",
+        output_processed_h5mu: "sample_ab_only_processed"
+      ]
+    ])
+    | map { state -> [state.id, state] }
+    | cellranger_multi_qc
+    | view { output ->
+      assert output.size() == 2 : "Outputs should contain two elements; [id, state]"
+
+      def id = output[0]
+      assert id == "sample_ab_only" : "Output ID should be 'sample_ab_only'. Found: ${id}"
+
+      def state = output[1]
+      assert state instanceof Map : "State should be a map. Found: ${state}"
+
+      assert state.containsKey("output_raw") : "State should contain key 'output_raw'."
+      assert state.output_raw.isDirectory() : "'output_raw' should be a directory."
+
+      assert state.containsKey("output_h5mu") : "State should contain key 'output_h5mu'."
+      assert state.output_h5mu.isFile() : "'output_h5mu' should be a file."
+
+      assert !state.containsKey("output_ingestion_qc_report") : "State should NOT contain 'output_ingestion_qc_report' for AB-only input."
+      assert !state.containsKey("output_multiqc_report") : "State should NOT contain 'output_multiqc_report' for AB-only input."
+
+      "Output: $output"
+    }
+}
+
+workflow test_wf_both_reports {
+  resources_test = file(params.resources_test)
+
+  output_ch = Channel.fromList([
+      [
+        id: "sample_both_reports",
+        input: [
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_GEX_1_subset_S1_L001_R1_001.fastq.gz"),
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_GEX_1_subset_S1_L001_R2_001.fastq.gz"),
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_AB_subset_S2_L004_R1_001.fastq.gz"),
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_AB_subset_S2_L004_R2_001.fastq.gz")
+        ],
+        gex_reference: resources_test.resolve("reference_gencodev41_chr1/reference_cellranger.tar.gz"),
+        feature_reference: resources_test.resolve("10x_5k_anticmv/raw/feature_reference.csv"),
+        library_id: ["5k_human_antiCMV_T_TBNK_connect_GEX_1_subset", "5k_human_antiCMV_T_TBNK_connect_AB_subset"],
+        library_type: ["Gene Expression", "Antibody Capture"],
+        output_raw: "sample_both_reports_raw/",
+        output_h5mu: "sample_both_reports.h5mu",
+        create_sample_qc_report: true,
+        create_multiqc_report: true,
+        output_ingestion_qc_report: "sample_both_reports_qc_report_*.html",
+        output_processed_h5mu: "sample_both_reports_processed"
+      ]
+    ])
+    | map { state -> [state.id, state] }
+    | cellranger_multi_qc
+    | view { output ->
+      assert output.size() == 2 : "Outputs should contain two elements; [id, state]"
+
+      def id = output[0]
+      assert id == "combined" : "Output ID should be 'combined'. Found: ${id}"
+
+      def state = output[1]
+      assert state instanceof Map : "State should be a map. Found: ${state}"
+
+      assert state.containsKey("output_raw") : "State should contain key 'output_raw'."
+      assert state.output_raw.isDirectory() : "'output_raw' should be a directory."
+
+      assert state.containsKey("output_h5mu") : "State should contain key 'output_h5mu'."
+      assert state.output_h5mu.isFile() : "'output_h5mu' should be a file."
+      assert state.output_h5mu.toString().endsWith(".h5mu") : "output_h5mu should end with '.h5mu'."
+
+      assert state.containsKey("output_multiqc_report") : "State should contain key 'output_multiqc_report'."
+      assert state.output_multiqc_report.isFile() : "'output_multiqc_report' should be a file."
+
+      assert state.containsKey("output_ingestion_qc_report") : "State should contain key 'output_ingestion_qc_report'."
+      assert state.output_ingestion_qc_report instanceof List : "'output_ingestion_qc_report' should be a list."
+      assert state.output_ingestion_qc_report.every { it.isFile() } : "All QC report files should exist."
+
+      assert state.containsKey("output_processed_h5mu") : "State should contain key 'output_processed_h5mu'."
+      assert state.output_processed_h5mu.isDirectory() : "'output_processed_h5mu' should be a directory."
+
+      "Output: $output"
+    }
+}
+
+workflow test_wf_multiqc_only {
+  resources_test = file(params.resources_test)
+
+  output_ch = Channel.fromList([
+      [
+        id: "sample_multiqc_only",
+        input: [
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_GEX_1_subset_S1_L001_R1_001.fastq.gz"),
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_GEX_1_subset_S1_L001_R2_001.fastq.gz"),
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_AB_subset_S2_L004_R1_001.fastq.gz"),
+          resources_test.resolve("10x_5k_anticmv/raw/5k_human_antiCMV_T_TBNK_connect_AB_subset_S2_L004_R2_001.fastq.gz")
+        ],
+        gex_reference: resources_test.resolve("reference_gencodev41_chr1/reference_cellranger.tar.gz"),
+        feature_reference: resources_test.resolve("10x_5k_anticmv/raw/feature_reference.csv"),
+        library_id: ["5k_human_antiCMV_T_TBNK_connect_GEX_1_subset", "5k_human_antiCMV_T_TBNK_connect_AB_subset"],
+        library_type: ["Gene Expression", "Antibody Capture"],
+        output_raw: "sample_multiqc_only_raw/",
+        output_h5mu: "sample_multiqc_only.h5mu",
+        create_sample_qc_report: false,
+        create_multiqc_report: true
+      ]
+    ])
+    | map { state -> [state.id, state] }
+    | cellranger_multi_qc
+    | view { output ->
+      assert output.size() == 2 : "Outputs should contain two elements; [id, state]"
+
+      def id = output[0]
+      assert id == "sample_multiqc_only" : "Output ID should be 'sample_multiqc_only'. Found: ${id}"
+
+      def state = output[1]
+      assert state instanceof Map : "State should be a map. Found: ${state}"
+
+      assert state.containsKey("output_raw") : "State should contain key 'output_raw'."
+      assert state.output_raw.isDirectory() : "'output_raw' should be a directory."
+
+      assert state.containsKey("output_h5mu") : "State should contain key 'output_h5mu'."
+      assert state.output_h5mu.isFile() : "'output_h5mu' should be a file."
+      assert state.output_h5mu.toString().endsWith(".h5mu") : "output_h5mu should end with '.h5mu'."
+
+      assert state.containsKey("output_multiqc_report") : "State should contain key 'output_multiqc_report'."
+      assert state.output_multiqc_report.isFile() : "'output_multiqc_report' should be a file."
+
+      assert !state.containsKey("output_ingestion_qc_report") : "State should NOT contain 'output_ingestion_qc_report' when only MultiQC is enabled."
+      assert !state.containsKey("output_processed_h5mu") : "State should NOT contain 'output_processed_h5mu' when only MultiQC is enabled."
+
+      "Output: $output"
+    }
+}
+
diff --git a/src/single_cell/process_integrate_annotate/config.vsh.yaml b/src/single_cell/process_integrate_annotate/config.vsh.yaml
new file mode 100644
index 0000000..748ef01
--- /dev/null
+++ b/src/single_cell/process_integrate_annotate/config.vsh.yaml
@@ -0,0 +1,386 @@
+name: "process_integrate_annotate"
+# scope: private
+namespace: "single_cell"
+description: |
+  A pipeline to process, integrate and annotate single cell (multi-)omics data.
+  Available integration methods:
+  - Harmony
+  - scVI
+  Available annotation methods:
+  - CellTypist
+  - scANVI (with scArches)
+
+authors:
+  - __merge__: /src/authors/dorien_roosen.yaml
+    roles: [ author, maintainer ]
+  - __merge__: /src/authors/weiwei_schultz.yaml
+    roles: [ contributor ]
+
+argument_groups:
+  - name: Input (query) data arguments
+    description: The input query dataset(s) to be annotated
+    arguments:
+      - name: "--id"
+        required: true
+        type: string
+        description: ID of the sample.
+        example: foo
+      - name: "--input"
+        required: true
+        type: file
+        description: Input query dataset(s) to be annotated
+        example: input.h5mu
+      - name: "--modality"
+        default: "rna"
+        type: string
+        description: Modality to be processed. Should match the modality in the --reference dataset, if provided.
+      - name: "--input_layer"
+        type: string
+        description: "The layer in the input data containing the raw counts, if .X is not to be used."
+        required: false
+      - name: "--input_var_gene_names"
+        type: string
+        required: false
+        description: |
+          The name of the adata var column containing gene names; when no gene_name_layer is provided, the var index will be used.
+      - name: "--input_reference_gene_overlap"
+        type: integer
+        default: 100
+        min: 1
+        description: | 
+          The minimum number of genes present in both the reference and query datasets.
+
+  - name: Reference data arguments
+    description: Dataset to be used as a reference for label transfer and to train annotation algorithms on.
+    arguments:
+      - name: "--reference"
+        type: file
+        required: false
+        example: reference.h5mu
+        description: |
+          The reference dataset in .h5mu format to be used as a reference mapper and to train annotation algorithms on.
+      - name: "--reference_layer_raw_counts"
+        type: string
+        description: "The layer in the reference dataset containing the raw counts, if .X is not to be used."
+        required: false
+      - name: "--reference_layer_lognormalized_counts"
+        type: string
+        default: log_normalized
+        description: "The layer in the reference dataset containing the log-normalized counts, if .X is not to be used."
+      - name: "--reference_var_gene_names"
+        type: string
+        required: false
+        description: |
+          The name of the adata .var column containing gene names if the .var index is not to be used.
+      - name: "--reference_obs_batch"
+        type: string
+        required: false 
+        description: |
+          The .obs column of the reference dataset containing the batch information.
+      - name: "--reference_obs_label"
+        type: string
+        example: cell_type
+        required: false
+        description: The `.obs` key of the target labels to tranfer.
+      - name: "--reference_obs_label_unlabeled_category"
+        type: string
+        default: "Unkown"
+        description: "Value in the --reference_obs_label field that indicates unlabeled observations"
+      - name: "--reference_var_input"
+        type: string
+        required: false
+        description: |
+          .var column containing highly variable genes. By default, do not subset genes.
+
+
+  - name: Methods
+    description: The available annotation and integration methods to integrate and/or annotate the query dataset(s) with.
+    arguments:
+      - name: "--integration_methods"
+        type: string
+        multiple: true
+        required: false
+        choices: [harmony, scvi]
+        example: harmony;scvi
+        description: Integration methods to be executed.
+      - name: "--annotation_methods"
+        type: string
+        multiple: true
+        required: false
+        choices: [celltypist, scanvi_scarches]
+        example: celltypist;scanvi_scarches
+        description: Annotation methods to be executed.
+        
+  - name: "Pre-processing options: RNA filtering"
+    description: Pre-processing options for filtering RNA data
+    arguments:
+      - name: "--rna_min_counts"
+        example: 200
+        type: integer
+        description: Minimum number of counts captured per cell.
+      - name: "--rna_max_counts"
+        example: 5000000
+        type: integer
+        description: Maximum number of counts captured per cell.
+      - name: "--rna_min_genes_per_cell"
+        type: integer
+        example: 200
+        description: Minimum of non-zero values per cell.
+      - name: "--rna_max_genes_per_cell"
+        example: 1500000
+        type: integer
+        description: Maximum of non-zero values per cell.
+      - name: "--rna_min_cells_per_gene"
+        example: 3
+        type: integer
+        description: Minimum of non-zero values per gene.
+      - name: "--rna_min_fraction_mito"
+        example: 0
+        type: double
+        description: Minimum fraction of UMIs that are mitochondrial.
+      - name: "--rna_max_fraction_mito"
+        type: double
+        example: 0.2
+        description: Maximum fraction of UMIs that are mitochondrial.
+
+  - name: "Pre-processing options: Highly variable features detection"
+    description: Pre-processing options for detecting highly variable features
+    arguments:
+      - name: "--n_hvg"
+        type: integer
+        description: |
+          Number of highly-variable features to keep. 
+          Only relevant if HVG need to be calculated across query and reference datasets (e.g. for --annotation_methods scvi_knn and harmony_knn). 
+          For reference mapping-based methods, the HVG's specified in --reference_var_input will be used.
+        default: 2000
+  
+  - name: "Pre-processing options: Mitochondrial & Ribosomal Gene Detection"
+    description: Pre-processing options for detecting mitochondrial genes
+    arguments:
+      - name: "--var_name_mitochondrial_genes"
+        type: string
+        required: false
+        description: |
+          In which .var slot to store a boolean array corresponding the mitochondrial genes.
+      - name: "--var_name_ribosomal_genes"
+        type: string
+        required: false
+        description: |
+          In which .var slot to store a boolean array corresponding the ribosomal genes.
+      - name: "--obs_name_mitochondrial_fraction"
+        type: string
+        required: false
+        description: |
+          When specified, write the fraction of counts originating from mitochondrial genes 
+          (based on --mitochondrial_gene_regex) to an .obs column with the specified name.
+          Requires --var_name_mitochondrial_genes.
+      - name: "--obs_name_ribosomal_fraction"
+        type: string
+        required: false
+        description: |
+          When specified, write the fraction of counts originating from ribosomal genes 
+          (based on --ribosomal_gene_regex) to an .obs column with the specified name.
+          Requires --var_name_ribosomal_genes.
+      - name: --mitochondrial_gene_regex
+        type: string
+        description: |
+          Regex string that identifies mitochondrial genes from --var_gene_names.
+          By default will detect human and mouse mitochondrial genes from a gene symbol.
+        required: false
+        default: "^[mM][tT]-"
+      - name: --ribosomal_gene_regex
+        type: string
+        description: |
+          Regex string that identifies ribosomal genes from --var_gene_names.
+          By default will detect human and mouse ribosomal genes from a gene symbol.
+        required: false
+        default: "^[Mm]?[Rr][Pp][LlSs]"
+
+  - name: "Pre-processing options: QC metrics calculation options"
+    description: Pre-processing options for calculating QC metrics
+    arguments:
+      - name: "--var_qc_metrics"
+        description: |
+          Keys to select a boolean (containing only True or False) column from .var.
+          For each cell, calculate the proportion of total values for genes which are labeled 'True', 
+          compared to the total sum of the values for all genes. Defaults to the combined values specified for
+          --var_name_mitochondrial_genes and --highly_variable_features_var_output.
+        type: string
+        multiple: True
+        multiple_sep: ','
+        required: false
+        example: "ercc,highly_variable"
+
+  - name: Harmony integration options
+    description: Specifications for harmony integration.
+    arguments:
+      - name: "--harmony_theta"
+        type: double
+        description: |
+          Diversity clustering penalty parameter. Specify for each variable in group.by.vars. 
+          theta=0 does not encourage any diversity. Larger values of theta
+          result in more diverse clusters."
+        default: 2
+        example: [0, 1, 2]
+        multiple: true
+      - name: "--harmony_obs_covariates"
+        type: string
+        description: "The .obs field(s) that define the covariate(s) to regress out."
+        example: ["batch", "sample"]
+        required: true
+        multiple: true
+        default: "sample_id"
+
+  - name: scVI, scANVI and scArches training options
+    # TODO - possibly provide separate training options for scVI, scANVI and scArches
+    description: Training arguments for scVI, scANVI and scArches. Relevant for --annotation_methods 'scvi_knn' and 'scanvi_scarches'.
+    arguments:
+      - name: "--early_stopping"
+        required: false
+        type: boolean
+        description: "Whether to perform early stopping with respect to the validation set."
+      - name: "--early_stopping_monitor"
+        choices: ["elbo_validation", "reconstruction_loss_validation", "kl_local_validation"]
+        default: "elbo_validation"
+        type: string
+        description: "Metric logged during validation set epoch."
+      - name: "--early_stopping_patience"
+        type: integer
+        min: 1
+        default: 45
+        description: "Number of validation epochs with no improvement after which training will be stopped."
+      - name: "--early_stopping_min_delta"
+        min: 0
+        type: double
+        default: 0.0
+        description: "Minimum change in the monitored quantity to qualify as an improvement, i.e. an absolute change of less than min_delta, will count as no improvement."
+      - name: "--max_epochs"
+        type: integer
+        description: "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest."
+        required: false
+      - name: "--reduce_lr_on_plateau"
+        description: "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus."
+        type: boolean
+        default: True
+      - name: "--lr_factor"
+        description: "Factor to reduce learning rate."
+        type: double
+        default: 0.6
+        min: 0
+      - name: "--lr_patience"
+        description: "Number of epochs with no improvement after which learning rate will be reduced."
+        type: double
+        default: 30
+        min: 0
+
+  - name: CellTypist reference model
+    description: The CellTypist reference model to use for annotation. If not provided, the reference dataset will be used for model training.
+    arguments:
+      - name: "--celltypist_model"
+        type: file
+        description: "Pretrained model in pkl format. If not provided, the model will be trained on the reference data and --reference should be provided."
+        required: false
+        example: pretrained_model.pkl
+
+  - name: CellTypist annotation options
+    description: Specifications for CellTypist annotation.
+    arguments:
+      - name: "--celltypist_feature_selection"
+        type: boolean
+        description: "Whether to perform feature selection."
+        default: false
+      - name: "--celltypist_majority_voting"
+        type: boolean
+        description: "Whether to refine the predicted labels by running the majority voting classifier after over-clustering."
+        default: false
+      - name: "--celltypist_C"
+        type: double
+        description: "Inverse of regularization strength in logistic regression."
+        default: 1.0
+      - name: "--celltypist_max_iter"
+        type: integer
+        description: "Maximum number of iterations before reaching the minimum of the cost function."
+        default: 1000
+      - name: "--celltypist_use_SGD"
+        type: boolean_true
+        description: "Whether to use the stochastic gradient descent algorithm."
+      - name: "--celltypist_min_prop"
+        type: double
+        description: |
+          "For the dominant cell type within a subcluster, the minimum proportion of cells required to 
+          support naming of the subcluster by this cell type. Ignored if majority_voting is set to False. 
+          Subcluster that fails to pass this proportion threshold will be assigned 'Heterogeneous'."
+        default: 0
+
+  - name: Clustering options
+    description: Arguments for Leiden clustering. Only relevant for --annotation_methods `scvi_knn`, `scanvi_scarches` and `harmony_knn`.
+    arguments:
+      - name: "--leiden_resolution"
+        type: double
+        description: Control the coarseness of the clustering. Higher values lead to more clusters.
+        default: [1]
+        multiple: true
+
+  - name: Neighbor classifier arguments
+    description: Arguments related to calculating the n nearest neighbors. Only relevant for --annotation_methods `scvi_knn`, `scanvi_scarches` and `harmony_knn`.
+    arguments:
+      - name: "--knn_weights"
+        type: string
+        default: "uniform"
+        choices: ["uniform", "distance"]
+        description: |
+          Weight function used in prediction. Possible values are:
+          `uniform` (all points in each neighborhood are weighted equally) or 
+          `distance` (weight points by the inverse of their distance)
+      - name: "--knn_n_neighbors"
+        type: integer
+        default: 15
+        min: 5
+        required: false
+        description: |
+          The number of neighbors to use in k-neighbor graph structure used for fast approximate nearest neighbor search with PyNNDescent. 
+          Larger values will result in more accurate search results at the cost of computation time.
+
+  - name: Outputs
+    description: The output file to write the annotated dataset to.
+    arguments:
+    - name: "--output"
+      type: file
+      direction: output
+      required: true
+      description: |
+        The output file.
+      example: output.h5mu
+
+dependencies:
+  - name: workflows/multiomics/process_samples
+    alias: process_samples_workflow
+    repository: openpipeline
+  - name: annotate/celltypist
+    repository: openpipeline
+    alias: celltypist_annotation
+  - name: workflows/annotation/scanvi_scarches
+    repository: openpipeline
+    alias: scanvi_scarches_annotation
+  - name: workflows/integration/harmony_leiden
+    repository: openpipeline
+    alias: harmony_integration
+  - name: workflows/integration/scvi_leiden
+    repository: openpipeline
+    alias: scvi_integration
+
+resources:
+  - type: nextflow_script
+    path: main.nf
+    entrypoint: run_wf
+
+test_resources:
+  - type: nextflow_script
+    path: test.nf
+    entrypoint: test_wf
+  - path: /resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_mms.h5mu
+  - path: /resources_test/annotation_test_data/TS_Blood_filtered.h5mu
+  - path: /resources_test/annotation_test_data/celltypist_model_Immune_All_Low.pkl
+
+runners:
+  - type: nextflow
\ No newline at end of file
diff --git a/src/single_cell/process_integrate_annotate/integration_test.sh b/src/single_cell/process_integrate_annotate/integration_test.sh
new file mode 100755
index 0000000..4659c0a
--- /dev/null
+++ b/src/single_cell/process_integrate_annotate/integration_test.sh
@@ -0,0 +1,37 @@
+#!/bin/bash
+
+set -eo pipefail
+
+# get the root of the directory
+REPO_ROOT=$(git rev-parse --show-toplevel)
+
+# ensure that the command below is run from the root of the repository
+cd "$REPO_ROOT"
+
+nextflow \
+  run . \
+  -main-script src/single_cell/process_integrate_annotate/test.nf \
+  -entry test_wf \
+  -resume \
+  -profile docker \
+  -c src/configs/labels_ci.config \
+  -c src/configs/integration_tests.config \
+  --publish_dir test
+
+nextflow \
+  run . \
+  -main-script src/single_cell/process_integrate_annotate/test.nf \
+  -profile docker,no_publish \
+  -resume \
+  -entry test_wf_2 \
+  -c src/configs/labels_ci.config \
+  -c src/configs/integration_tests.config
+
+nextflow \
+  run . \
+  -main-script src/single_cell/process_integrate_annotate/test.nf \
+  -profile docker,no_publish \
+  -resume \
+  -entry test_wf_3 \
+  -c src/configs/labels_ci.config \
+  -c src/configs/integration_tests.config
diff --git a/src/single_cell/process_integrate_annotate/main.nf b/src/single_cell/process_integrate_annotate/main.nf
new file mode 100644
index 0000000..5e795da
--- /dev/null
+++ b/src/single_cell/process_integrate_annotate/main.nf
@@ -0,0 +1,210 @@
+workflow run_wf {
+  take:
+    input_ch
+
+  main:
+    output_ch = input_ch
+    | map { id, state ->
+      def new_state = state + [ "query_processed": state.output, "_meta": ["join_id": id] ]
+      [id, new_state]
+    }
+    // Make sure parameters are filled out correctly
+    | map { id, state ->
+      def new_state = [:]
+      // Check that at least one of annotation_methods or integration_methods is not empty
+      if (!state.annotation_methods  && !state.integration_methods) {
+        throw new RuntimeException("At least one of --annotation_methods or --integration_methods must be provided")
+      }
+      // Check CellTypist arguments
+      if (state.annotation_methods && state.annotation_methods.contains("celltypist") && 
+        (!state.celltypist_model && !state.reference)) {
+        throw new RuntimeException("Celltypist was selected as an annotation method. Either --celltypist_model or --reference must be provided.")
+      }
+      if (state.annotation_methods && state.annotation_methods.contains("celltypist") && state.celltypist_model && state.reference )  {
+        System.err.println(
+          "Warning: --celltypist_model is set and a --reference was provided. \
+          The pre-trained Celltypist model will be used for annotation, the reference will be ignored."
+        )
+      }
+
+      [id, state + new_state]
+    }
+    | process_samples_workflow.run(
+      fromState: [
+        "input": "input", 
+        "id": "id",
+        "rna_layer": "input_layer",
+        "rna_min_counts": "rna_min_counts",
+        "rna_max_counts": "rna_max_counts",
+        "rna_min_genes_per_cell": "rna_min_genes_per_cell",
+        "rna_max_genes_per_cell": "rna_max_genes_per_cell",
+        "rna_min_cells_per_gene": "rna_min_cells_per_gene",
+        "rna_min_fraction_mito": "rna_min_fraction_mito",
+        "rna_max_fraction_mito": "rna_max_fraction_mito",
+        "rna_min_fraction_ribo": "rna_min_fraction_ribo",
+        "rna_max_fraction_ribo": "rna_max_fraction_ribo",
+        "var_name_mitochondrial_genes": "var_name_mitochondrial_genes",
+        "var_name_ribosomal_genes": "var_name_ribosomal_genes",
+        "var_gene_names": "input_var_gene_names",
+        "mitochondrial_gene_regex": "mitochondrial_gene_regex",
+        "ribosomal_gene_regex": "ribosomal_gene_regex",
+        "var_qc_metrics": "var_qc_metrics"
+      ],
+      args: [
+        "pca_overwrite": "true",
+        "add_id_obs_output": "sample_id",
+        "highly_variable_features_var_output": "filter_with_hvg_query"
+      ],
+      toState: ["query_processed": "output"], 
+    )
+    // Integration methods
+    | harmony_integration.run(
+      runIf: { id, state -> 
+        state.integration_methods && state.integration_methods.contains("harmony") 
+      },
+      fromState: [ 
+        "id": "id",
+        "input": "query_processed",
+        "modality": "modality",
+        "theta": "harmony_theta",
+        "leiden_resolution": "leiden_resolution",
+        "obs_covariates": "harmony_obs_covariates"
+      ],
+      args: [
+        "layer": "log_normalized",
+        "embedding": "X_pca",
+        "obsm_integrated": "X_harmony_integrated",
+        "uns_neighbors": "harmony_integration_neighbors",
+        "obsp_neighbor_distances": "harmony_integration_neighbor_distances",
+        "obsp_neighbor_connectivities": "harmony_integration_neighbor_connectivities",
+        "obs_cluster": "harmony_integration_leiden",
+        "obsm_umap": "X_harmony_umap"
+      ],
+      toState: [ "query_processed": "output" ]
+    )
+
+    | scvi_integration.run(
+      runIf: { id, state -> 
+        state.integration_methods && state.integration_methods.contains("scvi")
+      },
+      fromState: [ 
+        "id": "id",
+        "input": "query_processed",
+        "layer": "input_layer",
+        "modality": "modality",
+        "leiden_resolution": "leiden_resolution",
+        "early_stopping": "early_stopping",
+        "early_stopping_monitor": "early_stopping_monitor",
+        "early_stopping_patience": "early_stopping_patience",
+        "early_stopping_min_delta": "early_stopping_min_delta",
+        "max_epochs": "max_epochs",
+        "reduce_lr_on_plateau": "reduce_lr_on_plateau",
+        "lr_factor": "lr_factor",
+        "lr_patience": "lr_patience"
+      ],
+      args: [
+        "obsm_output": "X_scvi_integrated",
+        "obs_batch": "sample_id",
+        "var_input": "filter_with_hvg_query",
+        "uns_neighbors": "scvi_integration_neighbors",
+        "obsp_neighbor_distances": "scvi_integration_neighbor_distances",
+        "obsp_neighbor_connectivities": "scvi_integration_neighbor_connectivities",
+        "obs_cluster": "scvi_integration_leiden",
+        "obsm_umap": "X_scvi_umap"
+      ],
+      toState: [ "query_processed": "output", "scvi_model": "output_model" ]
+    )
+
+    // Annotation methods
+    | celltypist_annotation.run(
+      runIf: { id, state -> state.annotation_methods && state.annotation_methods.contains("celltypist") && state.celltypist_model },
+      fromState: [ 
+        "input": "query_processed",
+        "modality": "modality",
+        "input_var_gene_names": "input_var_gene_names",
+        "input_reference_gene_overlap": "input_reference_gene_overlap",
+        "model": "celltypist_model",
+        "majority_voting": "celltypist_majority_voting"
+      ],
+      args: [
+        // log normalized counts are expected for celltypist
+        "input_layer": "log_normalized",
+        "output_obs_predictions": "celltypist_pred",
+        "output_obs_probability": "celltypist_proba"
+      ],
+      toState: [ "query_processed": "output" ]
+    )
+
+    | celltypist_annotation.run(
+      runIf: { id, state -> state.annotation_methods && state.annotation_methods.contains("celltypist") && !state.celltypist_model },
+      fromState: [
+        "input": "query_processed",
+        "modality": "modality",
+        "input_var_gene_names": "input_var_gene_names",
+        "input_reference_gene_overlap": "input_reference_gene_overlap",
+        "reference": "reference",
+        "reference_layer": "reference_layer_lognormalized_counts",
+        "reference_obs_target": "reference_obs_label",
+        "reference_var_gene_names": "reference_var_gene_names",
+        "reference_obs_batch": "reference_obs_batch",
+        "reference_var_input": "reference_var_input",
+        "feature_selection": "celltypist_feature_selection",
+        "C": "celltypist_C",
+        "max_iter": "celltypist_max_iter",
+        "use_SGD": "celltypist_use_SGD",
+        "min_prop": "celltypist_min_prop",
+        "majority_voting": "celltypist_majority_voting"
+      ],
+      args: [
+        // log normalized counts are expected for celltypist
+        "input_layer": "log_normalized",
+        "output_obs_predictions": "celltypist_pred",
+        "output_obs_probability": "celltypist_proba"
+      ],
+      toState: [ "query_processed": "output" ]
+    )
+
+    | scanvi_scarches_annotation.run(
+      runIf: { id, state -> state.annotation_methods && state.annotation_methods.contains("scanvi_scarches")},
+      fromState: [
+        "id": "id",
+        "input": "query_processed",
+        "modality": "modality",
+        "layer": "input_layer",
+        "input_var_gene_names": "input_var_gene_names",
+        "reference": "reference",
+        "reference_obs_target": "reference_obs_label",
+        "reference_obs_batch_label": "reference_obs_batch",
+        "reference_var_hvg": "reference_var_input",
+        "reference_var_gene_names": "reference_var_gene_names",
+        "unlabeled_category": "reference_obs_label_unlabeled_category",
+        "early_stopping": "early_stopping",
+        "early_stopping_monitor": "early_stopping_monitor",
+        "early_stopping_patience": "early_stopping_patience",
+        "early_stopping_min_delta": "early_stopping_min_delta",
+        "max_epochs": "max_epochs",
+        "reduce_lr_on_plateau": "reduce_lr_on_plateau",
+        "lr_factor": "lr_factor",
+        "lr_patience": "lr_patience",
+        "leiden_resolution": "leiden_resolution",
+        "knn_weights": "knn_weights",
+        "knn_n_neighbors": "knn_n_neighbors"
+      ],
+      args: [
+        "input_obs_batch_label": "sample_id",
+        "output_obs_predictions": "scanvi_knn_pred",
+        "output_obs_probability": "scanvi_knn_proba"
+      ],
+      toState: [ "query_processed": "output" ]
+    )
+
+    | map {id, state ->
+      def new_state = state + ["output": state.query_processed]
+      [id, new_state]
+    }
+
+    | setState(["output", "_meta"])
+
+  emit:
+    output_ch
+}
\ No newline at end of file
diff --git a/src/single_cell/process_integrate_annotate/nextflow.config b/src/single_cell/process_integrate_annotate/nextflow.config
new file mode 100644
index 0000000..b0dc0d8
--- /dev/null
+++ b/src/single_cell/process_integrate_annotate/nextflow.config
@@ -0,0 +1,10 @@
+manifest {
+  nextflowVersion = '!>=20.12.1-edge'
+}
+
+params {
+  rootDir = java.nio.file.Paths.get("$projectDir/../../../").toAbsolutePath().normalize().toString()
+}
+
+// include common settings
+includeConfig("${params.rootDir}/src/configs/labels.config")
\ No newline at end of file
diff --git a/src/single_cell/process_integrate_annotate/test.nf b/src/single_cell/process_integrate_annotate/test.nf
new file mode 100644
index 0000000..b110015
--- /dev/null
+++ b/src/single_cell/process_integrate_annotate/test.nf
@@ -0,0 +1,151 @@
+nextflow.enable.dsl=2
+
+include { process_integrate_annotate } from params.rootDir + "/target/nextflow/single_cell/process_integrate_annotate/main.nf"
+params.resources_test = "s3://openpipelines-bio/openpipeline_incubator/resources_test/"
+
+workflow test_wf {
+  resources_test = file(params.resources_test)
+
+  output_ch = Channel.fromList(
+    [
+      [
+        id: "simple_annotation_test",
+        input: resources_test.resolve("pbmc_1k_protein_v3/pbmc_1k_protein_v3_mms.h5mu"),
+        reference: resources_test.resolve("annotation_test_data/TS_Blood_filtered.h5mu"),
+        reference_var_gene_names: "ensemblid",
+        reference_layer_lognormalized_counts: "log_normalized",
+        reference_obs_batch: "donor_assay",
+        reference_obs_label: "cell_type",
+        max_epochs: "5",
+        annotation_methods: "celltypist;scanvi_scarches"
+      ],
+      [
+        id: "simple_integration_test",
+        input: resources_test.resolve("pbmc_1k_protein_v3/pbmc_1k_protein_v3_mms.h5mu"),
+        integration_methods: "harmony;scvi"
+      ],
+      [
+        id: "simple_execution_test",
+        input: resources_test.resolve("pbmc_1k_protein_v3/pbmc_1k_protein_v3_mms.h5mu"),
+        reference: resources_test.resolve("annotation_test_data/TS_Blood_filtered.h5mu"),
+        reference_var_gene_names: "ensemblid",
+        reference_layer_lognormalized_counts: "log_normalized",
+        reference_obs_batch: "donor_assay",
+        reference_obs_label: "cell_type",
+        max_epochs: "5",
+        annotation_methods: "scanvi_scarches",
+        integration_methods: "harmony"
+      ]
+    ])
+    | view {"State at start: $it"}
+    | map{ state -> [state.id, state] }
+    | process_integrate_annotate 
+    | view {"After AaaS: $it"}
+    | view { output ->
+      assert output.size() == 2 : "Outputs should contain two elements; [id, state]"
+
+      // check id
+      def id = output[0]
+      assert id == "merged" : "Output ID should be `merged`"
+
+      // check output
+      def state = output[1]
+      assert state instanceof Map : "State should be a map. Found: ${state}"
+      assert state.containsKey("output") : "Output should contain key 'output'."
+      assert state.output.isFile() : "'output' should be a file."
+      assert state.output.toString().endsWith(".h5mu") : "Output file should end with '.h5mu'. Found: ${state.output}"
+    
+    "Output: $output"
+  }
+}
+
+workflow test_wf_2 {
+  resources_test = file(params.resources_test)
+
+  output_ch = Channel.fromList(
+    [
+      [
+        id: "pbmc_with_more_params",
+        input: resources_test.resolve("pbmc_1k_protein_v3/pbmc_1k_protein_v3_mms.h5mu"),
+        rna_min_counts: 2,
+        rna_max_counts: 1000000,
+        rna_min_genes_per_cell: 1,
+        rna_max_genes_per_cell: 1000000,
+        rna_min_cells_per_gene: 1,
+        rna_min_fraction_mito: 0.0,
+        rna_max_fraction_mito: 1.0,
+        prot_min_counts: 3,
+        prot_max_counts: 1000000,
+        prot_min_proteins_per_cell: 1,
+        prot_max_proteins_per_cell: 1000000,
+        prot_min_cells_per_protein: 1,
+        var_name_mitochondrial_genes: 'mitochondrial',
+        obs_name_mitochondrial_fraction: 'fraction_mitochondrial',
+        add_id_to_obs: true,
+        add_id_make_observation_keys_unique: true,
+        add_id_obs_output: "sample_id",
+        reference: resources_test.resolve("annotation_test_data/TS_Blood_filtered.h5mu"),
+        reference_var_gene_names: "ensemblid",
+        reference_layer_lognormalized_counts: "log_normalized",
+        reference_obs_batch: "donor_assay",
+        reference_obs_label: "cell_type",
+        annotation_methods: "celltypist",
+        integration_methods: "scvi"
+      ]
+    ])
+    | view {"State at start: $it"}
+    | map { state -> [state.id, state] }
+    | process_integrate_annotate 
+    | view {"After AaaS: $it"}
+    | view { output ->
+      assert output.size() == 2 : "Outputs should contain two elements; [id, state]"
+
+      // check id
+      def id = output[0]
+      assert id == "merged" : "Output ID should be `merged`"
+
+      // check output
+      def state = output[1]
+      assert state instanceof Map : "State should be a map. Found: ${state}"
+      assert state.containsKey("output") : "Output should contain key 'output'."
+      assert state.output.isFile() : "'output' should be a file."
+      assert state.output.toString().endsWith(".h5mu") : "Output file should end with '.h5mu'. Found: ${state.output}"
+    
+      "Output: $output"
+    }
+  }
+
+workflow test_wf_3 {
+  resources_test = file(params.resources_test)
+
+  output_ch = Channel.fromList(
+    [
+      [
+        id: "celltypist_model",
+        input: resources_test.resolve("pbmc_1k_protein_v3/pbmc_1k_protein_v3_mms.h5mu"),
+        celltypist_model: resources_test.resolve("annotation_test_data/celltypist_model_Immune_All_Low.pkl"),
+        annotation_methods: "celltypist",
+        input_var_gene_names: "gene_symbol"
+      ]
+    ])
+    | view {"State at start: $it"}
+    | map{ state -> [state.id, state] }
+    | process_integrate_annotate 
+    | view {"After AaaS: $it"}
+    | view { output ->
+      assert output.size() == 2 : "Outputs should contain two elements; [id, state]"
+
+      // check id
+      def id = output[0]
+      assert id == "merged" : "Output ID should be `merged`"
+
+      // check output
+      def state = output[1]
+      assert state instanceof Map : "State should be a map. Found: ${state}"
+      assert state.containsKey("output") : "Output should contain key 'output'."
+      assert state.output.isFile() : "'output' should be a file."
+      assert state.output.toString().endsWith(".h5mu") : "Output file should end with '.h5mu'. Found: ${state.output}"
+    
+    "Output: $output"
+  }
+}
diff --git a/target/.build.yaml b/target/.build.yaml
new file mode 100644
index 0000000..e69de29
diff --git a/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/.config.vsh.yaml b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/.config.vsh.yaml
new file mode 100644
index 0000000..1edfe64
--- /dev/null
+++ b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/.config.vsh.yaml
@@ -0,0 +1,388 @@
+name: "fastqc"
+version: "v0.4.2"
+authors:
+- name: "Theodoro Gasperin Terra Camargo"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    links:
+      email: "theodorogtc@gmail.com"
+      github: "tgaspe"
+      linkedin: "theodoro-gasperin-terra-camargo"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "FASTQ file(s) to be analyzed.\n"
+    info: null
+    example:
+    - "input.fq"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Outputs"
+  description: "At least one of the output options (--html, --zip, --summary, --data)\
+    \ must be used.\n"
+  arguments:
+  - type: "file"
+    name: "--outdir"
+    description: "Output directory where the results will be saved.\n"
+    info: null
+    example:
+    - "results"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--html"
+    description: "Create the HTML report of the results. \n'*' wild card must be provided\
+      \ in the output file name. \nWild card will be replaced by the input file basename.\n\
+      e.g. \n  --input \"sample_1.fq\"\n  --html \"*.html\"\n  would create an output\
+      \ html file named sample_1.html\n"
+    info: null
+    example:
+    - "*.html"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--zip"
+    description: "Create the zip file(s) containing: html report, data, images, icons,\
+      \ summary, etc.\n'*' wild card must be provided in the output file name.\nWild\
+      \ card will be replaced by the input basename.\ne.g. \n  --input \"sample_1.fq\"\
+      \n  --html \"*.zip\"\n  would create an output zip file named sample_1.zip\n"
+    info: null
+    example:
+    - "*.zip"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--summary"
+    description: "Create the summary file(s).\n'*' wild card must be provided in the\
+      \ output file name.\nWild card will be replaced by the input basename.\ne.g.\
+      \ \n  --input \"sample_1.fq\"\n  --summary \"*_summary.txt\"\n  would create\
+      \ an output summary.txt file named sample_1_summary.txt\n"
+    info: null
+    example:
+    - "*_summary.txt"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--data"
+    description: "Create the data file(s).\n'*' wild card must be provided in the\
+      \ output file name.\nWild card will be replaced by the input basename.\ne.g.\
+      \ \n  --input \"sample_1.fq\"\n  --summary \"*_data.txt\"\n  would create an\
+      \ output data.txt file named sample_1_data.txt\n"
+    info: null
+    example:
+    - "*_data.txt"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: true
+    multiple_sep: ";"
+- name: "Options"
+  arguments:
+  - type: "boolean_true"
+    name: "--casava"
+    description: "Files come from raw casava output. Files in the same sample\ngroup\
+      \ (differing only by the group number) will be analysed\nas a set rather than\
+      \ individually. Sequences with the filter\nflag set in the header will be excluded\
+      \ from the analysis.\nFiles must have the same names given to them by casava\n\
+      (including being gzipped and ending with .gz) otherwise they\nwon't be grouped\
+      \ together correctly.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--nano"
+    description: "Files come from nanopore sequences and are in fast5 format. In\n\
+      this mode you can pass in directories to process and the program\nwill take\
+      \ in all fast5 files within those directories and produce\na single output file\
+      \ from the sequences found in all files.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--nofilter"
+    description: "If running with --casava then don't remove read flagged by\ncasava\
+      \ as poor quality when performing the QC analysis.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--nogroup"
+    description: "Disable grouping of bases for reads >50bp. \nAll reports will show\
+      \ data for every base in the read. \nWARNING: Using this option will cause fastqc\
+      \ to crash \nand burn if you use it on really long reads, and your \nplots may\
+      \ end up a ridiculous size. You have been warned!\n"
+    info: null
+    direction: "input"
+  - type: "integer"
+    name: "--min_length"
+    description: "Sets an artificial lower limit on the length of the \nsequence to\
+      \ be shown in the report. As long as you \nset this to a value greater or equal\
+      \ to your longest \nread length then this will be the sequence length used \n\
+      to create your read groups. This can be useful for making\ndirectly comparable\
+      \ statistics from datasets with somewhat \nvariable read lengths.\n"
+    info: null
+    example:
+    - 0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--format"
+    alternatives:
+    - "-f"
+    description: "Bypasses the normal sequence file format detection and \nforces\
+      \ the program to use the specified format. \nValid formats are bam, sam, bam_mapped,\
+      \ sam_mapped, and fastq.\n"
+    info: null
+    example:
+    - "bam"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--contaminants"
+    alternatives:
+    - "-c"
+    description: "Specifies a non-default file which contains the list \nof contaminants\
+      \ to screen overrepresented sequences against. \nThe file must contain sets\
+      \ of named contaminants in the form\nname[tab]sequence. Lines prefixed with\
+      \ a hash will be ignored.\n"
+    info: null
+    example:
+    - "contaminants.txt"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--adapters"
+    alternatives:
+    - "-a"
+    description: "Specifies a non-default file which contains the list of \nadapter\
+      \ sequences which will be explicitly searched against \nthe library. The file\
+      \ must contain sets of named adapters \nin the form name[tab]sequence. Lines\
+      \ prefixed with a hash will be ignored.\n"
+    info: null
+    example:
+    - "adapters.txt"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--limits"
+    alternatives:
+    - "-l"
+    description: "Specifies a non-default file which contains \na set of criteria\
+      \ which will be used to determine \nthe warn/error limits for the various modules.\
+      \ \nThis file can also be used to selectively remove \nsome modules from the\
+      \ output altogether. The format \nneeds to mirror the default limits.txt file\
+      \ found in \nthe Configuration folder.\n"
+    info: null
+    example:
+    - "limits.txt"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--kmers"
+    alternatives:
+    - "-k"
+    description: "Specifies the length of Kmer to look for in the Kmer \ncontent module.\
+      \ Specified Kmer length must be between \n2 and 10. Default length is 7 if not\
+      \ specified.\n"
+    info: null
+    example:
+    - 7
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--quiet"
+    alternatives:
+    - "-q"
+    description: "Suppress all progress messages on stdout and only report errors.\n"
+    info: null
+    direction: "input"
+resources:
+- type: "bash_script"
+  path: "script.sh"
+  is_executable: true
+description: "FastQC - A high throughput sequence QC analysis tool."
+test_resources:
+- type: "bash_script"
+  path: "test.sh"
+  is_executable: true
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+keywords:
+- "Quality control"
+- "BAM"
+- "SAM"
+- "FASTQ"
+license: "GPL-3.0, Apache-2.0"
+links:
+  repository: "https://github.com/s-andrews/FastQC"
+  homepage: "https://www.bioinformatics.babraham.ac.uk/projects/fastqc/"
+  documentation: "https://www.bioinformatics.babraham.ac.uk/projects/fastqc/Help/"
+  issue_tracker: "https://github.com/s-andrews/FastQC/issues"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "biocontainers/fastqc:v0.11.9_cv8"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v0.4.2"
+  namespace_separator: "/"
+  setup:
+  - type: "docker"
+    run:
+    - "echo \"fastqc: $(fastqc --version | sed -n 's/^FastQC //p')\" > /var/software_versions.txt\n"
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/fastqc/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/fastqc"
+  executable: "target/nextflow/fastqc/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "02b470d967226478af69c37eae2b1256be1b78fd"
+  git_remote: "https://github.com/viash-hub/biobox"
+package_config:
+  name: "biobox"
+  version: "v0.4.2"
+  summary: "A curated collection of high-quality, standalone bioinformatics components\
+    \ built with [Viash](https://viash.io).\n"
+  description: "`biobox` offers a suite of reliable bioinformatics components, similar\
+    \ to [nf-core/modules](https://github.com/nf-core/modules) and [snakemake-wrappers/bio](https://github.com/snakemake/snakemake-wrappers/tree/master/bio),\
+    \ but built using the [Viash](https://viash.io) framework.\n\nThis approach emphasizes\
+    \ **reusability**, **reproducibility**, and adherence to **best practices**. Key\
+    \ features of `biobox` components include:\n\n* **Standalone & Nextflow Ready:**\
+    \ Run components directly via the command line or seamlessly integrate them into\
+    \ Nextflow workflows.\n* **High Quality Standards:**\n    * Comprehensive documentation\
+    \ for components and parameters.\n    * Full exposure of underlying tool arguments.\n\
+    \    * Containerized (Docker) for dependency management and reproducibility.\n\
+    \    * Unit tested for verified functionality.\n"
+  info: null
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands += ['ps']\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v0.4.2'"
+  keywords:
+  - "bioinformatics"
+  - "modules"
+  - "sequencing"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/viash-hub/biobox"
+    issue_tracker: "https://github.com/viash-hub/biobox/issues"
diff --git a/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/main.nf b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/main.nf
new file mode 100644
index 0000000..36489f6
--- /dev/null
+++ b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/main.nf
@@ -0,0 +1,4111 @@
+// fastqc v0.4.2
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Theodoro Gasperin Terra Camargo (author, maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "fastqc",
+  "version" : "v0.4.2",
+  "authors" : [
+    {
+      "name" : "Theodoro Gasperin Terra Camargo",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "links" : {
+          "email" : "theodorogtc@gmail.com",
+          "github" : "tgaspe",
+          "linkedin" : "theodoro-gasperin-terra-camargo"
+        }
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "FASTQ file(s) to be analyzed.\n",
+          "example" : [
+            "input.fq"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "description" : "At least one of the output options (--html, --zip, --summary, --data) must be used.\n",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--outdir",
+          "description" : "Output directory where the results will be saved.\n",
+          "example" : [
+            "results"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--html",
+          "description" : "Create the HTML report of the results. \n'*' wild card must be provided in the output file name. \nWild card will be replaced by the input file basename.\ne.g. \n  --input \\"sample_1.fq\\"\n  --html \\"*.html\\"\n  would create an output html file named sample_1.html\n",
+          "example" : [
+            "*.html"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--zip",
+          "description" : "Create the zip file(s) containing: html report, data, images, icons, summary, etc.\n'*' wild card must be provided in the output file name.\nWild card will be replaced by the input basename.\ne.g. \n  --input \\"sample_1.fq\\"\n  --html \\"*.zip\\"\n  would create an output zip file named sample_1.zip\n",
+          "example" : [
+            "*.zip"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--summary",
+          "description" : "Create the summary file(s).\n'*' wild card must be provided in the output file name.\nWild card will be replaced by the input basename.\ne.g. \n  --input \\"sample_1.fq\\"\n  --summary \\"*_summary.txt\\"\n  would create an output summary.txt file named sample_1_summary.txt\n",
+          "example" : [
+            "*_summary.txt"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--data",
+          "description" : "Create the data file(s).\n'*' wild card must be provided in the output file name.\nWild card will be replaced by the input basename.\ne.g. \n  --input \\"sample_1.fq\\"\n  --summary \\"*_data.txt\\"\n  would create an output data.txt file named sample_1_data.txt\n",
+          "example" : [
+            "*_data.txt"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Options",
+      "arguments" : [
+        {
+          "type" : "boolean_true",
+          "name" : "--casava",
+          "description" : "Files come from raw casava output. Files in the same sample\ngroup (differing only by the group number) will be analysed\nas a set rather than individually. Sequences with the filter\nflag set in the header will be excluded from the analysis.\nFiles must have the same names given to them by casava\n(including being gzipped and ending with .gz) otherwise they\nwon't be grouped together correctly.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--nano",
+          "description" : "Files come from nanopore sequences and are in fast5 format. In\nthis mode you can pass in directories to process and the program\nwill take in all fast5 files within those directories and produce\na single output file from the sequences found in all files.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--nofilter",
+          "description" : "If running with --casava then don't remove read flagged by\ncasava as poor quality when performing the QC analysis.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--nogroup",
+          "description" : "Disable grouping of bases for reads >50bp. \nAll reports will show data for every base in the read. \nWARNING: Using this option will cause fastqc to crash \nand burn if you use it on really long reads, and your \nplots may end up a ridiculous size. You have been warned!\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_length",
+          "description" : "Sets an artificial lower limit on the length of the \nsequence to be shown in the report. As long as you \nset this to a value greater or equal to your longest \nread length then this will be the sequence length used \nto create your read groups. This can be useful for making\ndirectly comparable statistics from datasets with somewhat \nvariable read lengths.\n",
+          "example" : [
+            0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--format",
+          "alternatives" : [
+            "-f"
+          ],
+          "description" : "Bypasses the normal sequence file format detection and \nforces the program to use the specified format. \nValid formats are bam, sam, bam_mapped, sam_mapped, and fastq.\n",
+          "example" : [
+            "bam"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--contaminants",
+          "alternatives" : [
+            "-c"
+          ],
+          "description" : "Specifies a non-default file which contains the list \nof contaminants to screen overrepresented sequences against. \nThe file must contain sets of named contaminants in the form\nname[tab]sequence. Lines prefixed with a hash will be ignored.\n",
+          "example" : [
+            "contaminants.txt"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--adapters",
+          "alternatives" : [
+            "-a"
+          ],
+          "description" : "Specifies a non-default file which contains the list of \nadapter sequences which will be explicitly searched against \nthe library. The file must contain sets of named adapters \nin the form name[tab]sequence. Lines prefixed with a hash will be ignored.\n",
+          "example" : [
+            "adapters.txt"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--limits",
+          "alternatives" : [
+            "-l"
+          ],
+          "description" : "Specifies a non-default file which contains \na set of criteria which will be used to determine \nthe warn/error limits for the various modules. \nThis file can also be used to selectively remove \nsome modules from the output altogether. The format \nneeds to mirror the default limits.txt file found in \nthe Configuration folder.\n",
+          "example" : [
+            "limits.txt"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--kmers",
+          "alternatives" : [
+            "-k"
+          ],
+          "description" : "Specifies the length of Kmer to look for in the Kmer \ncontent module. Specified Kmer length must be between \n2 and 10. Default length is 7 if not specified.\n",
+          "example" : [
+            7
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--quiet",
+          "alternatives" : [
+            "-q"
+          ],
+          "description" : "Suppress all progress messages on stdout and only report errors.\n",
+          "direction" : "input"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "bash_script",
+      "path" : "script.sh",
+      "is_executable" : true
+    }
+  ],
+  "description" : "FastQC - A high throughput sequence QC analysis tool.",
+  "test_resources" : [
+    {
+      "type" : "bash_script",
+      "path" : "test.sh",
+      "is_executable" : true
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "keywords" : [
+    "Quality control",
+    "BAM",
+    "SAM",
+    "FASTQ"
+  ],
+  "license" : "GPL-3.0, Apache-2.0",
+  "links" : {
+    "repository" : "https://github.com/s-andrews/FastQC",
+    "homepage" : "https://www.bioinformatics.babraham.ac.uk/projects/fastqc/",
+    "documentation" : "https://www.bioinformatics.babraham.ac.uk/projects/fastqc/Help/",
+    "issue_tracker" : "https://github.com/s-andrews/FastQC/issues"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        }
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "biocontainers/fastqc:v0.11.9_cv8",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v0.4.2",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "docker",
+          "run" : [
+            "echo \\"fastqc: $(fastqc --version | sed -n 's/^FastQC //p')\\" > /var/software_versions.txt\n"
+          ]
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/fastqc/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "target/nextflow/fastqc",
+    "viash_version" : "0.9.4",
+    "git_commit" : "02b470d967226478af69c37eae2b1256be1b78fd",
+    "git_remote" : "https://github.com/viash-hub/biobox"
+  },
+  "package_config" : {
+    "name" : "biobox",
+    "version" : "v0.4.2",
+    "summary" : "A curated collection of high-quality, standalone bioinformatics components built with [Viash](https://viash.io).\n",
+    "description" : "`biobox` offers a suite of reliable bioinformatics components, similar to [nf-core/modules](https://github.com/nf-core/modules) and [snakemake-wrappers/bio](https://github.com/snakemake/snakemake-wrappers/tree/master/bio), but built using the [Viash](https://viash.io) framework.\n\nThis approach emphasizes **reusability**, **reproducibility**, and adherence to **best practices**. Key features of `biobox` components include:\n\n* **Standalone & Nextflow Ready:** Run components directly via the command line or seamlessly integrate them into Nextflow workflows.\n* **High Quality Standards:**\n    * Comprehensive documentation for components and parameters.\n    * Full exposure of underlying tool arguments.\n    * Containerized (Docker) for dependency management and reproducibility.\n    * Unit tested for verified functionality.\n",
+    "viash_version" : "0.9.4",
+    "source" : "src",
+    "target" : "target",
+    "config_mods" : [
+      ".requirements.commands += ['ps']\n",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v0.4.2'"
+    ],
+    "keywords" : [
+      "bioinformatics",
+      "modules",
+      "sequencing"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/viash-hub/biobox",
+      "issue_tracker" : "https://github.com/viash-hub/biobox/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.sh"
+cat > "$tempscript" << VIASHMAIN
+#!/bin/bash
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+$( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "${VIASH_PAR_INPUT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_input='&'#" ; else echo "# par_input="; fi )
+$( if [ ! -z ${VIASH_PAR_OUTDIR+x} ]; then echo "${VIASH_PAR_OUTDIR}" | sed "s#'#'\\"'\\"'#g;s#.*#par_outdir='&'#" ; else echo "# par_outdir="; fi )
+$( if [ ! -z ${VIASH_PAR_HTML+x} ]; then echo "${VIASH_PAR_HTML}" | sed "s#'#'\\"'\\"'#g;s#.*#par_html='&'#" ; else echo "# par_html="; fi )
+$( if [ ! -z ${VIASH_PAR_ZIP+x} ]; then echo "${VIASH_PAR_ZIP}" | sed "s#'#'\\"'\\"'#g;s#.*#par_zip='&'#" ; else echo "# par_zip="; fi )
+$( if [ ! -z ${VIASH_PAR_SUMMARY+x} ]; then echo "${VIASH_PAR_SUMMARY}" | sed "s#'#'\\"'\\"'#g;s#.*#par_summary='&'#" ; else echo "# par_summary="; fi )
+$( if [ ! -z ${VIASH_PAR_DATA+x} ]; then echo "${VIASH_PAR_DATA}" | sed "s#'#'\\"'\\"'#g;s#.*#par_data='&'#" ; else echo "# par_data="; fi )
+$( if [ ! -z ${VIASH_PAR_CASAVA+x} ]; then echo "${VIASH_PAR_CASAVA}" | sed "s#'#'\\"'\\"'#g;s#.*#par_casava='&'#" ; else echo "# par_casava="; fi )
+$( if [ ! -z ${VIASH_PAR_NANO+x} ]; then echo "${VIASH_PAR_NANO}" | sed "s#'#'\\"'\\"'#g;s#.*#par_nano='&'#" ; else echo "# par_nano="; fi )
+$( if [ ! -z ${VIASH_PAR_NOFILTER+x} ]; then echo "${VIASH_PAR_NOFILTER}" | sed "s#'#'\\"'\\"'#g;s#.*#par_nofilter='&'#" ; else echo "# par_nofilter="; fi )
+$( if [ ! -z ${VIASH_PAR_NOGROUP+x} ]; then echo "${VIASH_PAR_NOGROUP}" | sed "s#'#'\\"'\\"'#g;s#.*#par_nogroup='&'#" ; else echo "# par_nogroup="; fi )
+$( if [ ! -z ${VIASH_PAR_MIN_LENGTH+x} ]; then echo "${VIASH_PAR_MIN_LENGTH}" | sed "s#'#'\\"'\\"'#g;s#.*#par_min_length='&'#" ; else echo "# par_min_length="; fi )
+$( if [ ! -z ${VIASH_PAR_FORMAT+x} ]; then echo "${VIASH_PAR_FORMAT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_format='&'#" ; else echo "# par_format="; fi )
+$( if [ ! -z ${VIASH_PAR_CONTAMINANTS+x} ]; then echo "${VIASH_PAR_CONTAMINANTS}" | sed "s#'#'\\"'\\"'#g;s#.*#par_contaminants='&'#" ; else echo "# par_contaminants="; fi )
+$( if [ ! -z ${VIASH_PAR_ADAPTERS+x} ]; then echo "${VIASH_PAR_ADAPTERS}" | sed "s#'#'\\"'\\"'#g;s#.*#par_adapters='&'#" ; else echo "# par_adapters="; fi )
+$( if [ ! -z ${VIASH_PAR_LIMITS+x} ]; then echo "${VIASH_PAR_LIMITS}" | sed "s#'#'\\"'\\"'#g;s#.*#par_limits='&'#" ; else echo "# par_limits="; fi )
+$( if [ ! -z ${VIASH_PAR_KMERS+x} ]; then echo "${VIASH_PAR_KMERS}" | sed "s#'#'\\"'\\"'#g;s#.*#par_kmers='&'#" ; else echo "# par_kmers="; fi )
+$( if [ ! -z ${VIASH_PAR_QUIET+x} ]; then echo "${VIASH_PAR_QUIET}" | sed "s#'#'\\"'\\"'#g;s#.*#par_quiet='&'#" ; else echo "# par_quiet="; fi )
+$( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "${VIASH_META_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_name='&'#" ; else echo "# meta_name="; fi )
+$( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "${VIASH_META_FUNCTIONALITY_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_functionality_name='&'#" ; else echo "# meta_functionality_name="; fi )
+$( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "${VIASH_META_RESOURCES_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_resources_dir='&'#" ; else echo "# meta_resources_dir="; fi )
+$( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "${VIASH_META_EXECUTABLE}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_executable='&'#" ; else echo "# meta_executable="; fi )
+$( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "${VIASH_META_CONFIG}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_config='&'#" ; else echo "# meta_config="; fi )
+$( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "${VIASH_META_TEMP_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_temp_dir='&'#" ; else echo "# meta_temp_dir="; fi )
+$( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "${VIASH_META_CPUS}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_cpus='&'#" ; else echo "# meta_cpus="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "${VIASH_META_MEMORY_B}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_b='&'#" ; else echo "# meta_memory_b="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "${VIASH_META_MEMORY_KB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_kb='&'#" ; else echo "# meta_memory_kb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "${VIASH_META_MEMORY_MB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_mb='&'#" ; else echo "# meta_memory_mb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "${VIASH_META_MEMORY_GB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_gb='&'#" ; else echo "# meta_memory_gb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "${VIASH_META_MEMORY_TB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_tb='&'#" ; else echo "# meta_memory_tb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "${VIASH_META_MEMORY_PB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_pb='&'#" ; else echo "# meta_memory_pb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "${VIASH_META_MEMORY_KIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_kib='&'#" ; else echo "# meta_memory_kib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "${VIASH_META_MEMORY_MIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_mib='&'#" ; else echo "# meta_memory_mib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "${VIASH_META_MEMORY_GIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_gib='&'#" ; else echo "# meta_memory_gib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "${VIASH_META_MEMORY_TIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_tib='&'#" ; else echo "# meta_memory_tib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "${VIASH_META_MEMORY_PIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_pib='&'#" ; else echo "# meta_memory_pib="; fi )
+
+## VIASH END
+
+# exit on error
+set -eo pipefail
+
+# Check if both outputs are empty, at least one must be passed.
+if [[ -z "\\$par_outdir" ]] && [[ -z "\\$par_html" ]] && [[ -z "\\$par_zip" ]] && [[ -z "\\$par_summary" ]] && [[ -z "\\$par_data" ]]; then
+  echo "Error: At least one of the output arguments (--outdir, --html, --zip, --summary, and --data) must be passed."
+  exit 1
+fi
+
+# unset flags
+unset_if_false=(
+  par_casava
+  par_nano
+  par_nofilter
+  par_extract
+  par_noextract
+  par_nogroup
+  par_quiet
+)
+
+for par in \\${unset_if_false[@]}; do
+    test_val="\\${!par}"
+    [[ "\\$test_val" == "false" ]] && unset \\$par
+done
+
+tmpdir=\\$(mktemp -d "\\${meta_temp_dir}/\\${meta_name}-XXXXXXXX")
+function clean_up {
+  rm -rf "\\$tmpdir"
+}
+trap clean_up EXIT
+
+# Set output directory
+if [[ -n "\\$par_outdir" ]]; then
+  if [[ ! -d "\\$par_outdir" ]]; then
+    mkdir -p "\\$par_outdir"
+  fi
+  output_dir="\\$par_outdir"
+else
+  output_dir="\\$tmpdir"
+fi
+
+# Create input array 
+IFS=";" read -ra input <<< \\$par_input
+
+# Run fastqc
+fastqc \\\\
+  --extract \\\\
+  \\${par_casava:+--casava} \\\\
+  \\${par_nano:+--nano} \\\\
+  \\${par_nofilter:+--nofilter} \\\\
+  \\${par_nogroup:+--nogroup} \\\\
+  \\${par_min_length:+--min_length "\\$par_min_length"} \\\\
+  \\${par_format:+--format "\\$par_format"} \\\\
+  \\${par_contaminants:+--contaminants "\\$par_contaminants"} \\\\
+  \\${par_adapters:+--adapters "\\$par_adapters"} \\\\
+  \\${par_limits:+--limits "\\$par_limits"} \\\\
+  \\${par_kmers:+--kmers "\\$par_kmers"} \\\\
+  \\${par_quiet:+--quiet} \\\\
+  \\${meta_cpus:+--threads "\\$meta_cpus"} \\\\
+  \\${meta_temp_dir:+--dir "\\$meta_temp_dir"} \\\\
+  --outdir "\\${output_dir}" \\\\
+  "\\${input[@]}"
+ 
+
+# Move output files
+for file in "\\${input[@]}"; do
+  # Removes everything after the first dot of the basename
+  sample_name=\\$(basename "\\${file}" | sed 's/\\\\..*\\$//')
+  if [[ -n "\\$par_html" ]]; then
+    input_html="\\${output_dir}/\\${sample_name}_fastqc.html"
+    if [[ ! -f "\\$input_html" ]]; then
+      echo "WARNING: HTML file '\\$input_html' does not exist"
+    else
+      html_file="\\${par_html//\\\\*/\\$sample_name}"
+      cp "\\$input_html" "\\$html_file"
+    fi
+  fi
+  if [[ -n "\\$par_zip" ]]; then
+    input_zip="\\${output_dir}/\\${sample_name}_fastqc.zip"
+    if [[ ! -f "\\$input_zip" ]]; then
+      echo "WARNING: ZIP file '\\$input_zip' does not exist"
+    else
+      zip_file="\\${par_zip//\\\\*/\\$sample_name}"
+      cp "\\$input_zip" "\\$zip_file"
+    fi
+  fi
+  if [[ -n "\\$par_summary" ]]; then
+    summary_file="\\${output_dir}/\\${sample_name}_fastqc/summary.txt"
+    if [[ ! -f "\\$summary_file" ]]; then
+      echo "WARNING: Summary file '\\$summary_file' does not exist"
+    else
+      new_summary="\\${par_summary//\\\\*/\\$sample_name}"
+      cp "\\$summary_file" "\\$new_summary"
+    fi
+  fi
+  if [[ -n "\\$par_data" ]]; then
+    data_file="\\${output_dir}/\\${sample_name}_fastqc/fastqc_data.txt"
+    if [[ ! -f "\\$data_file" ]]; then
+      echo "WARNING: Data file '\\$data_file' does not exist"
+    else
+      new_data="\\${par_data//\\\\*/\\$sample_name}"
+      cp "\\$data_file" "\\$new_data"
+    fi
+  fi
+done
+VIASHMAIN
+bash "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/biobox/fastqc",
+    "tag" : "v0.4.2"
+  },
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/nextflow.config b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/nextflow.config
new file mode 100644
index 0000000..21ed65f
--- /dev/null
+++ b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'fastqc'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v0.4.2'
+  description = 'FastQC - A high throughput sequence QC analysis tool.'
+  author = 'Theodoro Gasperin Terra Camargo'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+
diff --git a/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/nextflow_schema.json b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/nextflow_schema.json
new file mode 100644
index 0000000..ca2d0ce
--- /dev/null
+++ b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/nextflow_schema.json
@@ -0,0 +1,175 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "fastqc",
+  "description": "FastQC - A high throughput sequence QC analysis tool.",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "exists": true,
+          "description": "FASTQ file(s) to be analyzed.\n",
+          "help_text": "Type: `file`, multiple: `True`, required, direction: `input`, example: `[\"input.fq\"]`. "
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "At least one of the output options (--html, --zip, --summary, --data) must be used.\n",
+      "properties": {
+        "outdir": {
+          "type": "string",
+          "format": "path",
+          "description": "Output directory where the results will be saved.\n",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.outdir\"`, direction: `output`, example: `\"results\"`. ",
+          "default": "$id.$key.outdir"
+        },
+        "html": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "Create the HTML report of the results",
+          "help_text": "Type: `file`, multiple: `True`, default: `\"$id.$key.html_*.html\"`, direction: `output`, example: `[\"*.html\"]`. ",
+          "default": "$id.$key.html_*.html"
+        },
+        "zip": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "Create the zip file(s) containing: html report, data, images, icons, summary, etc.\n'*' wild card must be provided in the output file name.\nWild card will be replaced by the input basename.\ne.g",
+          "help_text": "Type: `file`, multiple: `True`, default: `\"$id.$key.zip_*.zip\"`, direction: `output`, example: `[\"*.zip\"]`. ",
+          "default": "$id.$key.zip_*.zip"
+        },
+        "summary": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "Create the summary file(s).\n'*' wild card must be provided in the output file name.\nWild card will be replaced by the input basename.\ne.g",
+          "help_text": "Type: `file`, multiple: `True`, default: `\"$id.$key.summary_*.txt\"`, direction: `output`, example: `[\"*_summary.txt\"]`. ",
+          "default": "$id.$key.summary_*.txt"
+        },
+        "data": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "Create the data file(s).\n'*' wild card must be provided in the output file name.\nWild card will be replaced by the input basename.\ne.g",
+          "help_text": "Type: `file`, multiple: `True`, default: `\"$id.$key.data_*.txt\"`, direction: `output`, example: `[\"*_data.txt\"]`. ",
+          "default": "$id.$key.data_*.txt"
+        }
+      }
+    },
+    "options": {
+      "title": "Options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "casava": {
+          "type": "boolean",
+          "description": "Files come from raw casava output",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "nano": {
+          "type": "boolean",
+          "description": "Files come from nanopore sequences and are in fast5 format",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "nofilter": {
+          "type": "boolean",
+          "description": "If running with --casava then don't remove read flagged by\ncasava as poor quality when performing the QC analysis.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "nogroup": {
+          "type": "boolean",
+          "description": "Disable grouping of bases for reads >50bp",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "min_length": {
+          "type": "integer",
+          "description": "Sets an artificial lower limit on the length of the \nsequence to be shown in the report",
+          "help_text": "Type: `integer`, multiple: `False`, example: `0`. "
+        },
+        "format": {
+          "type": "string",
+          "description": "Bypasses the normal sequence file format detection and \nforces the program to use the specified format",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"bam\"`. "
+        },
+        "contaminants": {
+          "type": "string",
+          "format": "path",
+          "description": "Specifies a non-default file which contains the list \nof contaminants to screen overrepresented sequences against",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"contaminants.txt\"`. "
+        },
+        "adapters": {
+          "type": "string",
+          "format": "path",
+          "description": "Specifies a non-default file which contains the list of \nadapter sequences which will be explicitly searched against \nthe library",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"adapters.txt\"`. "
+        },
+        "limits": {
+          "type": "string",
+          "format": "path",
+          "description": "Specifies a non-default file which contains \na set of criteria which will be used to determine \nthe warn/error limits for the various modules",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"limits.txt\"`. "
+        },
+        "kmers": {
+          "type": "integer",
+          "description": "Specifies the length of Kmer to look for in the Kmer \ncontent module",
+          "help_text": "Type: `integer`, multiple: `False`, example: `7`. "
+        },
+        "quiet": {
+          "type": "boolean",
+          "description": "Suppress all progress messages on stdout and only report errors.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/.config.vsh.yaml b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/.config.vsh.yaml
new file mode 100644
index 0000000..0e0455d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/.config.vsh.yaml
@@ -0,0 +1,496 @@
+name: "multiqc"
+version: "v0.4.2"
+authors:
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Input"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "File paths to be searched for analysis results to be included in\
+      \ the report.\n"
+    info: null
+    example:
+    - "data/results"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Ouput"
+  arguments:
+  - type: "file"
+    name: "--output_report"
+    description: "Filepath of the generated report.\n"
+    info: null
+    example:
+    - "multiqc_report.html"
+    must_exist: false
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_data"
+    description: "Output directory for parsed data files. If not provided, parsed\
+      \ data will not be published.\n"
+    info: null
+    example:
+    - "multiqc_data"
+    must_exist: false
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_plots"
+    description: "Output directory for generated plots. If not provided, plots will\
+      \ not be published.\n"
+    info: null
+    example:
+    - "multiqc_plots"
+    must_exist: false
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Modules and analyses to run"
+  arguments:
+  - type: "string"
+    name: "--include_modules"
+    description: "Use only these module"
+    info: null
+    example:
+    - "fastqc"
+    - "cutadapt"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--exclude_modules"
+    description: "Do not use only these modules"
+    info: null
+    example:
+    - "fastqc"
+    - "cutadapt"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ignore_analysis"
+    info: null
+    example:
+    - "run_one/*"
+    - "run_two/*"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ignore_samples"
+    info: null
+    example:
+    - "sample_1*"
+    - "sample_3*"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--ignore_symlinks"
+    description: "Ignore symlinked directories and files"
+    info: null
+    direction: "input"
+- name: "Sample name handling"
+  arguments:
+  - type: "boolean_true"
+    name: "--dirs"
+    description: "Prepend directory to sample names to avoid clashing filenames"
+    info: null
+    direction: "input"
+  - type: "integer"
+    name: "--dirs_depth"
+    description: "Prepend n directories to sample names. Negative number to take from\
+      \ start of path."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--full_names"
+    description: "Do not clean the sample names (leave as full file name)"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--fn_as_s_name"
+    description: "Use the log filename as the sample name"
+    info: null
+    direction: "input"
+  - type: "file"
+    name: "--replace_names"
+    description: "TSV file to rename sample names during report generation"
+    info: null
+    example:
+    - "replace_names.tsv"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Report Customisation"
+  arguments:
+  - type: "string"
+    name: "--title"
+    description: "Report title. Printed as page header, used for filename if not otherwise\
+      \ specified.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--comment"
+    description: "Custom comment, will be printed at the top of the report.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--template"
+    description: "Report template to use.\n"
+    info: null
+    required: false
+    choices:
+    - "default"
+    - "gathered"
+    - "geo"
+    - "highcharts"
+    - "sections"
+    - "simple"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--sample_names"
+    description: "TSV file containing alternative sample names for renaming buttons\
+      \ in the report.\n"
+    info: null
+    example:
+    - "sample_names.tsv"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--sample_filters"
+    description: "TSV file containing show/hide patterns for the report\n"
+    info: null
+    example:
+    - "sample_filters.tsv"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--custom_css_file"
+    description: "Custom CSS file to add to the final report\n"
+    info: null
+    example:
+    - "custom_style_sheet.css"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--profile_runtime"
+    description: "Add analysis of how long MultiQC takes to run to the report\n"
+    info: null
+    direction: "input"
+- name: "MultiQC behaviour"
+  arguments:
+  - type: "boolean_true"
+    name: "--verbose"
+    description: "Increase output verbosity.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--quiet"
+    description: "Only show log warnings\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--strict"
+    description: "Don't catch exceptions, run additional code checks to help development.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--development"
+    description: "Development mode. Do not compress and minimise JS, export uncompressed\
+      \ plot data.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--require_logs"
+    description: "Require all explicitly requested modules to have log files. If not,\
+      \ MultiQC will exit with an error.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--no_megaqc_upload"
+    description: "Don't upload generated report to MegaQC, even if MegaQC options\
+      \ are found.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--no_ansi"
+    description: "Disable coloured log output.\n"
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--cl_config"
+    description: "YAML formatted string that allows to customize MultiQC behaviour\
+      \ like input file detection.\n"
+    info: null
+    example:
+    - "qualimap_config: { general_stats_coverage: [20,40,200] }"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Output format"
+  arguments:
+  - type: "boolean_true"
+    name: "--flat"
+    description: "Use only flat plots (static images).\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--interactive"
+    description: "Use only interactive plots (in-browser Javascript).\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--data_dir"
+    description: "Force the parsed data directory to be created.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--no_data_dir"
+    description: "Prevent the parsed data directory from being created.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--zip_data_dir"
+    description: "Compress the data directory.\n"
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--data_format"
+    description: "Output parsed data in a different format than the default 'txt'.\n"
+    info: null
+    required: false
+    choices:
+    - "tsv"
+    - "csv"
+    - "json"
+    - "yaml"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--pdf"
+    description: "Creates PDF report with the 'simple' template. Requires Pandoc to\
+      \ be installed.\n"
+    info: null
+    direction: "input"
+resources:
+- type: "bash_script"
+  path: "script.sh"
+  is_executable: true
+description: "MultiQC aggregates results from bioinformatics analyses across many\
+  \ samples into a single report.\nIt searches a given directory for analysis logs\
+  \ and compiles a HTML report. It's a general use tool, perfect for summarising the\
+  \ output from numerous bioinformatics tools.\n"
+test_resources:
+- type: "bash_script"
+  path: "test.sh"
+  is_executable: true
+- type: "file"
+  path: "test_data"
+info:
+  keywords:
+  - "QC"
+  - "html report"
+  - "aggregate analysis"
+  links:
+    homepage: "https://multiqc.info/"
+    documentation: "https://multiqc.info/docs/"
+    repository: "https://github.com/MultiQC/MultiQC"
+  references:
+    doi: "10.1093/bioinformatics/btw354"
+  licence: "GPL v3 or later"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+license: "MIT"
+links:
+  repository: "https://github.com/viash-hub/biobox"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "quay.io/biocontainers/multiqc:1.21--pyhdfd78af_0"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v0.4.2"
+  namespace_separator: "/"
+  setup:
+  - type: "docker"
+    run:
+    - "multiqc --version | sed 's/multiqc, version\\s\\(.*\\)/multiqc: \"\\1\"/' >\
+      \ /var/software_versions.txt\n"
+  test_setup:
+  - type: "apt"
+    packages:
+    - "jq"
+    interactive: false
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/multiqc/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/multiqc"
+  executable: "target/nextflow/multiqc/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "02b470d967226478af69c37eae2b1256be1b78fd"
+  git_remote: "https://github.com/viash-hub/biobox"
+package_config:
+  name: "biobox"
+  version: "v0.4.2"
+  summary: "A curated collection of high-quality, standalone bioinformatics components\
+    \ built with [Viash](https://viash.io).\n"
+  description: "`biobox` offers a suite of reliable bioinformatics components, similar\
+    \ to [nf-core/modules](https://github.com/nf-core/modules) and [snakemake-wrappers/bio](https://github.com/snakemake/snakemake-wrappers/tree/master/bio),\
+    \ but built using the [Viash](https://viash.io) framework.\n\nThis approach emphasizes\
+    \ **reusability**, **reproducibility**, and adherence to **best practices**. Key\
+    \ features of `biobox` components include:\n\n* **Standalone & Nextflow Ready:**\
+    \ Run components directly via the command line or seamlessly integrate them into\
+    \ Nextflow workflows.\n* **High Quality Standards:**\n    * Comprehensive documentation\
+    \ for components and parameters.\n    * Full exposure of underlying tool arguments.\n\
+    \    * Containerized (Docker) for dependency management and reproducibility.\n\
+    \    * Unit tested for verified functionality.\n"
+  info: null
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands += ['ps']\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v0.4.2'"
+  keywords:
+  - "bioinformatics"
+  - "modules"
+  - "sequencing"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/viash-hub/biobox"
+    issue_tracker: "https://github.com/viash-hub/biobox/issues"
diff --git a/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/main.nf b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/main.nf
new file mode 100644
index 0000000..a57400a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/main.nf
@@ -0,0 +1,4335 @@
+// multiqc v0.4.2
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dorien Roosen (author, maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "multiqc",
+  "version" : "v0.4.2",
+  "authors" : [
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Input",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "File paths to be searched for analysis results to be included in the report.\n",
+          "example" : [
+            "data/results"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Ouput",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output_report",
+          "description" : "Filepath of the generated report.\n",
+          "example" : [
+            "multiqc_report.html"
+          ],
+          "must_exist" : false,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_data",
+          "description" : "Output directory for parsed data files. If not provided, parsed data will not be published.\n",
+          "example" : [
+            "multiqc_data"
+          ],
+          "must_exist" : false,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_plots",
+          "description" : "Output directory for generated plots. If not provided, plots will not be published.\n",
+          "example" : [
+            "multiqc_plots"
+          ],
+          "must_exist" : false,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Modules and analyses to run",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--include_modules",
+          "description" : "Use only these module",
+          "example" : [
+            "fastqc",
+            "cutadapt"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--exclude_modules",
+          "description" : "Do not use only these modules",
+          "example" : [
+            "fastqc",
+            "cutadapt"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--ignore_analysis",
+          "example" : [
+            "run_one/*",
+            "run_two/*"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--ignore_samples",
+          "example" : [
+            "sample_1*",
+            "sample_3*"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--ignore_symlinks",
+          "description" : "Ignore symlinked directories and files",
+          "direction" : "input"
+        }
+      ]
+    },
+    {
+      "name" : "Sample name handling",
+      "arguments" : [
+        {
+          "type" : "boolean_true",
+          "name" : "--dirs",
+          "description" : "Prepend directory to sample names to avoid clashing filenames",
+          "direction" : "input"
+        },
+        {
+          "type" : "integer",
+          "name" : "--dirs_depth",
+          "description" : "Prepend n directories to sample names. Negative number to take from start of path.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--full_names",
+          "description" : "Do not clean the sample names (leave as full file name)",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--fn_as_s_name",
+          "description" : "Use the log filename as the sample name",
+          "direction" : "input"
+        },
+        {
+          "type" : "file",
+          "name" : "--replace_names",
+          "description" : "TSV file to rename sample names during report generation",
+          "example" : [
+            "replace_names.tsv"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Report Customisation",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--title",
+          "description" : "Report title. Printed as page header, used for filename if not otherwise specified.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--comment",
+          "description" : "Custom comment, will be printed at the top of the report.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--template",
+          "description" : "Report template to use.\n",
+          "required" : false,
+          "choices" : [
+            "default",
+            "gathered",
+            "geo",
+            "highcharts",
+            "sections",
+            "simple"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--sample_names",
+          "description" : "TSV file containing alternative sample names for renaming buttons in the report.\n",
+          "example" : [
+            "sample_names.tsv"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--sample_filters",
+          "description" : "TSV file containing show/hide patterns for the report\n",
+          "example" : [
+            "sample_filters.tsv"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--custom_css_file",
+          "description" : "Custom CSS file to add to the final report\n",
+          "example" : [
+            "custom_style_sheet.css"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--profile_runtime",
+          "description" : "Add analysis of how long MultiQC takes to run to the report\n",
+          "direction" : "input"
+        }
+      ]
+    },
+    {
+      "name" : "MultiQC behaviour",
+      "arguments" : [
+        {
+          "type" : "boolean_true",
+          "name" : "--verbose",
+          "description" : "Increase output verbosity.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--quiet",
+          "description" : "Only show log warnings\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--strict",
+          "description" : "Don't catch exceptions, run additional code checks to help development.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--development",
+          "description" : "Development mode. Do not compress and minimise JS, export uncompressed plot data.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--require_logs",
+          "description" : "Require all explicitly requested modules to have log files. If not, MultiQC will exit with an error.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--no_megaqc_upload",
+          "description" : "Don't upload generated report to MegaQC, even if MegaQC options are found.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--no_ansi",
+          "description" : "Disable coloured log output.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--cl_config",
+          "description" : "YAML formatted string that allows to customize MultiQC behaviour like input file detection.\n",
+          "example" : [
+            "qualimap_config: { general_stats_coverage: [20,40,200] }"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Output format",
+      "arguments" : [
+        {
+          "type" : "boolean_true",
+          "name" : "--flat",
+          "description" : "Use only flat plots (static images).\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--interactive",
+          "description" : "Use only interactive plots (in-browser Javascript).\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--data_dir",
+          "description" : "Force the parsed data directory to be created.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--no_data_dir",
+          "description" : "Prevent the parsed data directory from being created.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--zip_data_dir",
+          "description" : "Compress the data directory.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--data_format",
+          "description" : "Output parsed data in a different format than the default 'txt'.\n",
+          "required" : false,
+          "choices" : [
+            "tsv",
+            "csv",
+            "json",
+            "yaml"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--pdf",
+          "description" : "Creates PDF report with the 'simple' template. Requires Pandoc to be installed.\n",
+          "direction" : "input"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "bash_script",
+      "path" : "script.sh",
+      "is_executable" : true
+    }
+  ],
+  "description" : "MultiQC aggregates results from bioinformatics analyses across many samples into a single report.\nIt searches a given directory for analysis logs and compiles a HTML report. It's a general use tool, perfect for summarising the output from numerous bioinformatics tools.\n",
+  "test_resources" : [
+    {
+      "type" : "bash_script",
+      "path" : "test.sh",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "test_data"
+    }
+  ],
+  "info" : {
+    "keywords" : [
+      "QC",
+      "html report",
+      "aggregate analysis"
+    ],
+    "links" : {
+      "homepage" : "https://multiqc.info/",
+      "documentation" : "https://multiqc.info/docs/",
+      "repository" : "https://github.com/MultiQC/MultiQC"
+    },
+    "references" : {
+      "doi" : "10.1093/bioinformatics/btw354"
+    },
+    "licence" : "GPL v3 or later"
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/viash-hub/biobox"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        }
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "quay.io/biocontainers/multiqc:1.21--pyhdfd78af_0",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v0.4.2",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "docker",
+          "run" : [
+            "multiqc --version | sed 's/multiqc, version\\\\s\\\\(.*\\\\)/multiqc: \\"\\\\1\\"/' > /var/software_versions.txt\n"
+          ]
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "jq"
+          ],
+          "interactive" : false
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/multiqc/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "target/nextflow/multiqc",
+    "viash_version" : "0.9.4",
+    "git_commit" : "02b470d967226478af69c37eae2b1256be1b78fd",
+    "git_remote" : "https://github.com/viash-hub/biobox"
+  },
+  "package_config" : {
+    "name" : "biobox",
+    "version" : "v0.4.2",
+    "summary" : "A curated collection of high-quality, standalone bioinformatics components built with [Viash](https://viash.io).\n",
+    "description" : "`biobox` offers a suite of reliable bioinformatics components, similar to [nf-core/modules](https://github.com/nf-core/modules) and [snakemake-wrappers/bio](https://github.com/snakemake/snakemake-wrappers/tree/master/bio), but built using the [Viash](https://viash.io) framework.\n\nThis approach emphasizes **reusability**, **reproducibility**, and adherence to **best practices**. Key features of `biobox` components include:\n\n* **Standalone & Nextflow Ready:** Run components directly via the command line or seamlessly integrate them into Nextflow workflows.\n* **High Quality Standards:**\n    * Comprehensive documentation for components and parameters.\n    * Full exposure of underlying tool arguments.\n    * Containerized (Docker) for dependency management and reproducibility.\n    * Unit tested for verified functionality.\n",
+    "viash_version" : "0.9.4",
+    "source" : "src",
+    "target" : "target",
+    "config_mods" : [
+      ".requirements.commands += ['ps']\n",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v0.4.2'"
+    ],
+    "keywords" : [
+      "bioinformatics",
+      "modules",
+      "sequencing"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/viash-hub/biobox",
+      "issue_tracker" : "https://github.com/viash-hub/biobox/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.sh"
+cat > "$tempscript" << VIASHMAIN
+## VIASH START
+# The following code has been auto-generated by Viash.
+$( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "${VIASH_PAR_INPUT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_input='&'#" ; else echo "# par_input="; fi )
+$( if [ ! -z ${VIASH_PAR_OUTPUT_REPORT+x} ]; then echo "${VIASH_PAR_OUTPUT_REPORT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_output_report='&'#" ; else echo "# par_output_report="; fi )
+$( if [ ! -z ${VIASH_PAR_OUTPUT_DATA+x} ]; then echo "${VIASH_PAR_OUTPUT_DATA}" | sed "s#'#'\\"'\\"'#g;s#.*#par_output_data='&'#" ; else echo "# par_output_data="; fi )
+$( if [ ! -z ${VIASH_PAR_OUTPUT_PLOTS+x} ]; then echo "${VIASH_PAR_OUTPUT_PLOTS}" | sed "s#'#'\\"'\\"'#g;s#.*#par_output_plots='&'#" ; else echo "# par_output_plots="; fi )
+$( if [ ! -z ${VIASH_PAR_INCLUDE_MODULES+x} ]; then echo "${VIASH_PAR_INCLUDE_MODULES}" | sed "s#'#'\\"'\\"'#g;s#.*#par_include_modules='&'#" ; else echo "# par_include_modules="; fi )
+$( if [ ! -z ${VIASH_PAR_EXCLUDE_MODULES+x} ]; then echo "${VIASH_PAR_EXCLUDE_MODULES}" | sed "s#'#'\\"'\\"'#g;s#.*#par_exclude_modules='&'#" ; else echo "# par_exclude_modules="; fi )
+$( if [ ! -z ${VIASH_PAR_IGNORE_ANALYSIS+x} ]; then echo "${VIASH_PAR_IGNORE_ANALYSIS}" | sed "s#'#'\\"'\\"'#g;s#.*#par_ignore_analysis='&'#" ; else echo "# par_ignore_analysis="; fi )
+$( if [ ! -z ${VIASH_PAR_IGNORE_SAMPLES+x} ]; then echo "${VIASH_PAR_IGNORE_SAMPLES}" | sed "s#'#'\\"'\\"'#g;s#.*#par_ignore_samples='&'#" ; else echo "# par_ignore_samples="; fi )
+$( if [ ! -z ${VIASH_PAR_IGNORE_SYMLINKS+x} ]; then echo "${VIASH_PAR_IGNORE_SYMLINKS}" | sed "s#'#'\\"'\\"'#g;s#.*#par_ignore_symlinks='&'#" ; else echo "# par_ignore_symlinks="; fi )
+$( if [ ! -z ${VIASH_PAR_DIRS+x} ]; then echo "${VIASH_PAR_DIRS}" | sed "s#'#'\\"'\\"'#g;s#.*#par_dirs='&'#" ; else echo "# par_dirs="; fi )
+$( if [ ! -z ${VIASH_PAR_DIRS_DEPTH+x} ]; then echo "${VIASH_PAR_DIRS_DEPTH}" | sed "s#'#'\\"'\\"'#g;s#.*#par_dirs_depth='&'#" ; else echo "# par_dirs_depth="; fi )
+$( if [ ! -z ${VIASH_PAR_FULL_NAMES+x} ]; then echo "${VIASH_PAR_FULL_NAMES}" | sed "s#'#'\\"'\\"'#g;s#.*#par_full_names='&'#" ; else echo "# par_full_names="; fi )
+$( if [ ! -z ${VIASH_PAR_FN_AS_S_NAME+x} ]; then echo "${VIASH_PAR_FN_AS_S_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#par_fn_as_s_name='&'#" ; else echo "# par_fn_as_s_name="; fi )
+$( if [ ! -z ${VIASH_PAR_REPLACE_NAMES+x} ]; then echo "${VIASH_PAR_REPLACE_NAMES}" | sed "s#'#'\\"'\\"'#g;s#.*#par_replace_names='&'#" ; else echo "# par_replace_names="; fi )
+$( if [ ! -z ${VIASH_PAR_TITLE+x} ]; then echo "${VIASH_PAR_TITLE}" | sed "s#'#'\\"'\\"'#g;s#.*#par_title='&'#" ; else echo "# par_title="; fi )
+$( if [ ! -z ${VIASH_PAR_COMMENT+x} ]; then echo "${VIASH_PAR_COMMENT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_comment='&'#" ; else echo "# par_comment="; fi )
+$( if [ ! -z ${VIASH_PAR_TEMPLATE+x} ]; then echo "${VIASH_PAR_TEMPLATE}" | sed "s#'#'\\"'\\"'#g;s#.*#par_template='&'#" ; else echo "# par_template="; fi )
+$( if [ ! -z ${VIASH_PAR_SAMPLE_NAMES+x} ]; then echo "${VIASH_PAR_SAMPLE_NAMES}" | sed "s#'#'\\"'\\"'#g;s#.*#par_sample_names='&'#" ; else echo "# par_sample_names="; fi )
+$( if [ ! -z ${VIASH_PAR_SAMPLE_FILTERS+x} ]; then echo "${VIASH_PAR_SAMPLE_FILTERS}" | sed "s#'#'\\"'\\"'#g;s#.*#par_sample_filters='&'#" ; else echo "# par_sample_filters="; fi )
+$( if [ ! -z ${VIASH_PAR_CUSTOM_CSS_FILE+x} ]; then echo "${VIASH_PAR_CUSTOM_CSS_FILE}" | sed "s#'#'\\"'\\"'#g;s#.*#par_custom_css_file='&'#" ; else echo "# par_custom_css_file="; fi )
+$( if [ ! -z ${VIASH_PAR_PROFILE_RUNTIME+x} ]; then echo "${VIASH_PAR_PROFILE_RUNTIME}" | sed "s#'#'\\"'\\"'#g;s#.*#par_profile_runtime='&'#" ; else echo "# par_profile_runtime="; fi )
+$( if [ ! -z ${VIASH_PAR_VERBOSE+x} ]; then echo "${VIASH_PAR_VERBOSE}" | sed "s#'#'\\"'\\"'#g;s#.*#par_verbose='&'#" ; else echo "# par_verbose="; fi )
+$( if [ ! -z ${VIASH_PAR_QUIET+x} ]; then echo "${VIASH_PAR_QUIET}" | sed "s#'#'\\"'\\"'#g;s#.*#par_quiet='&'#" ; else echo "# par_quiet="; fi )
+$( if [ ! -z ${VIASH_PAR_STRICT+x} ]; then echo "${VIASH_PAR_STRICT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_strict='&'#" ; else echo "# par_strict="; fi )
+$( if [ ! -z ${VIASH_PAR_DEVELOPMENT+x} ]; then echo "${VIASH_PAR_DEVELOPMENT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_development='&'#" ; else echo "# par_development="; fi )
+$( if [ ! -z ${VIASH_PAR_REQUIRE_LOGS+x} ]; then echo "${VIASH_PAR_REQUIRE_LOGS}" | sed "s#'#'\\"'\\"'#g;s#.*#par_require_logs='&'#" ; else echo "# par_require_logs="; fi )
+$( if [ ! -z ${VIASH_PAR_NO_MEGAQC_UPLOAD+x} ]; then echo "${VIASH_PAR_NO_MEGAQC_UPLOAD}" | sed "s#'#'\\"'\\"'#g;s#.*#par_no_megaqc_upload='&'#" ; else echo "# par_no_megaqc_upload="; fi )
+$( if [ ! -z ${VIASH_PAR_NO_ANSI+x} ]; then echo "${VIASH_PAR_NO_ANSI}" | sed "s#'#'\\"'\\"'#g;s#.*#par_no_ansi='&'#" ; else echo "# par_no_ansi="; fi )
+$( if [ ! -z ${VIASH_PAR_CL_CONFIG+x} ]; then echo "${VIASH_PAR_CL_CONFIG}" | sed "s#'#'\\"'\\"'#g;s#.*#par_cl_config='&'#" ; else echo "# par_cl_config="; fi )
+$( if [ ! -z ${VIASH_PAR_FLAT+x} ]; then echo "${VIASH_PAR_FLAT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_flat='&'#" ; else echo "# par_flat="; fi )
+$( if [ ! -z ${VIASH_PAR_INTERACTIVE+x} ]; then echo "${VIASH_PAR_INTERACTIVE}" | sed "s#'#'\\"'\\"'#g;s#.*#par_interactive='&'#" ; else echo "# par_interactive="; fi )
+$( if [ ! -z ${VIASH_PAR_DATA_DIR+x} ]; then echo "${VIASH_PAR_DATA_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#par_data_dir='&'#" ; else echo "# par_data_dir="; fi )
+$( if [ ! -z ${VIASH_PAR_NO_DATA_DIR+x} ]; then echo "${VIASH_PAR_NO_DATA_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#par_no_data_dir='&'#" ; else echo "# par_no_data_dir="; fi )
+$( if [ ! -z ${VIASH_PAR_ZIP_DATA_DIR+x} ]; then echo "${VIASH_PAR_ZIP_DATA_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#par_zip_data_dir='&'#" ; else echo "# par_zip_data_dir="; fi )
+$( if [ ! -z ${VIASH_PAR_DATA_FORMAT+x} ]; then echo "${VIASH_PAR_DATA_FORMAT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_data_format='&'#" ; else echo "# par_data_format="; fi )
+$( if [ ! -z ${VIASH_PAR_PDF+x} ]; then echo "${VIASH_PAR_PDF}" | sed "s#'#'\\"'\\"'#g;s#.*#par_pdf='&'#" ; else echo "# par_pdf="; fi )
+$( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "${VIASH_META_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_name='&'#" ; else echo "# meta_name="; fi )
+$( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "${VIASH_META_FUNCTIONALITY_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_functionality_name='&'#" ; else echo "# meta_functionality_name="; fi )
+$( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "${VIASH_META_RESOURCES_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_resources_dir='&'#" ; else echo "# meta_resources_dir="; fi )
+$( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "${VIASH_META_EXECUTABLE}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_executable='&'#" ; else echo "# meta_executable="; fi )
+$( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "${VIASH_META_CONFIG}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_config='&'#" ; else echo "# meta_config="; fi )
+$( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "${VIASH_META_TEMP_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_temp_dir='&'#" ; else echo "# meta_temp_dir="; fi )
+$( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "${VIASH_META_CPUS}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_cpus='&'#" ; else echo "# meta_cpus="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "${VIASH_META_MEMORY_B}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_b='&'#" ; else echo "# meta_memory_b="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "${VIASH_META_MEMORY_KB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_kb='&'#" ; else echo "# meta_memory_kb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "${VIASH_META_MEMORY_MB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_mb='&'#" ; else echo "# meta_memory_mb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "${VIASH_META_MEMORY_GB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_gb='&'#" ; else echo "# meta_memory_gb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "${VIASH_META_MEMORY_TB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_tb='&'#" ; else echo "# meta_memory_tb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "${VIASH_META_MEMORY_PB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_pb='&'#" ; else echo "# meta_memory_pb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "${VIASH_META_MEMORY_KIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_kib='&'#" ; else echo "# meta_memory_kib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "${VIASH_META_MEMORY_MIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_mib='&'#" ; else echo "# meta_memory_mib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "${VIASH_META_MEMORY_GIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_gib='&'#" ; else echo "# meta_memory_gib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "${VIASH_META_MEMORY_TIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_tib='&'#" ; else echo "# meta_memory_tib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "${VIASH_META_MEMORY_PIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_pib='&'#" ; else echo "# meta_memory_pib="; fi )
+
+## VIASH END
+#!/bin/bash
+
+# disable flags
+unset_if_false=(
+    par_ignore_symlinks
+    par_dirs
+    par_full_names
+    par_fn_as_s_name
+    par_profile_runtime
+    par_verbose
+    par_quiet
+    par_strict
+    par_development
+    par_require_logs
+    par_no_megaqc_upload
+    par_no_ansi
+    par_flat
+    par_interactive
+    par_static_plot_export
+    par_data_dir
+    par_no_data_dir
+    par_zip_data_dir
+    par_pdf
+)
+
+for par in \\${unset_if_false[@]}; do
+    test_val="\\${!par}"
+    [[ "\\$test_val" == "false" ]] && unset \\$par
+done
+
+# handle inputs
+out_dir=\\$(dirname "\\$par_output_report")
+output_report_file=\\$(basename "\\$par_output_report")
+report_name="\\${output_report_file%.*}"
+
+# handle outputs
+[[ -z "\\$par_output_report" ]] && no_report=true
+[[ -z "\\$par_output_data" ]] && no_data_dir=true
+[[ ! -z "\\$par_output_data" ]] && data_dir=true
+[[ ! -z "\\$par_output_plots" ]] && export=true
+
+# handle multiples
+IFS=";" read -ra inputs <<< \\$par_input
+
+if [[ -n "\\$par_include_modules" ]]; then
+    include_modules=""
+    IFS=";" read -ra incl_modules <<< \\$par_include_modules
+    for i in "\\${incl_modules[@]}"; do
+        include_modules+="--include \\$i "
+    done
+    unset IFS
+fi
+
+if [[ -n "\\$par_exclude_modules" ]]; then
+    exclude_modules=""
+    IFS=";" read -ra excl_modules <<< \\$par_exclude_modules
+    for i in "\\${excl_modules[@]}"; do
+        exclude_modules+="--exclude \\$i"
+    done
+    unset IFS
+fi
+
+if [[ -n "\\$par_ignore_analysis" ]]; then
+    ignore=""
+    IFS=";" read -ra ignore_analysis <<< \\$par_ignore_analysis
+    for i in "\\${ignore_analysis[@]}"; do
+        ignore+="--ignore \\$i "
+    done
+    unset IFS
+fi
+
+if [[ -n "\\$par_ignore_samples" ]]; then
+    ignore_samples=""
+    IFS=";" read -ra ign_samples <<< \\$par_ignore_samples
+    for i in "\\${ign_samples[@]}"; do
+        ignore_samples+="--ignore-samples \\$i"
+    done
+    unset IFS
+fi
+
+# run multiqc
+multiqc \\\\
+    \\${par_output_report:+--filename "\\$report_name"} \\\\
+    \\${out_dir:+--outdir "\\$out_dir"} \\\\
+    \\${no_report:+--no-report} \\\\
+    \\${no_data_dir:+--no-data-dir} \\\\
+    \\${data_dir:+--data-dir} \\\\
+    \\${export:+--export} \\\\
+    \\${par_title:+--title "\\$par_title"} \\\\
+    \\${par_comment:+--comment "\\$par_comment"} \\\\
+    \\${par_template:+--template "\\$par_template"} \\\\
+    \\${par_sample_names:+--sample-names "\\$par_sample_names"} \\\\
+    \\${par_sample_filters:+--sample-filters "\\$par_sample_filters"} \\\\
+    \\${par_custom_css_file:+--custom-css-file "\\$par_custom_css_file"} \\\\
+    \\${par_profile_runtime:+--profile-runtime} \\\\
+    \\${par_dirs:+--dirs} \\\\
+    \\${par_dirs_depth:+--dirs-depth "\\$par_dirs_depth"} \\\\
+    \\${par_full_names:+--full-names} \\\\
+    \\${par_fn_as_s_name:+--fn-as-s-name} \\\\
+    \\${par_ignore_names:+--ignore-names "\\$par_ignore_names"} \\\\
+    \\${par_ignore_symlinks:+--ignore-symlinks} \\\\
+    \\${ignore_samples} \\\\
+    \\${ignore} \\\\
+    \\${exclude_modules} \\\\
+    \\${include_modules} \\\\
+    \\${par_include_modules:+--include-modules "\\$par_include_modules"} \\\\
+    \\${par_data_format:+--data-format "\\$par_data_format"} \\\\
+    \\${par_cl_config:+--cl-config "\\$par_cl_config"} \\\\
+    \\${par_zip_data_dir:+--zip-data-dir} \\\\
+    \\${par_pdf:+--pdf} \\\\
+    \\${par_interactive:+--interactive} \\\\
+    \\${par_flat:+--flat} \\\\
+    \\${par_verbose:+--verbose} \\\\
+    \\${par_quiet:+--quiet} \\\\
+    \\${par_strict:+--strict} \\\\
+    \\${par_no_megaqc_upload:+--no-megaqc-upload} \\\\
+    \\${par_no_ansi:+--no-ansi} \\\\
+    \\${par_profile_runtime:+--profile-runtime} \\\\
+    \\${par_require_logs:+--require-logs} \\\\
+    \\${par_development:+--development} \\\\
+    --force \\\\
+    "\\${inputs[@]}" 
+
+# Move outputs
+
+if [[ -n "\\$par_output_data" ]] && [[ -d "\\${out_dir}/\\${report_name}_data" ]]; then
+    mv "\\${out_dir}/\\${report_name}_data" "\\$par_output_data"
+elif [[ -n "\\$par_output_data" ]] && [[ ! -d "\\${out_dir}/\\${report_name}_data" ]]; then
+    echo "WARNING: Data could not be saved because data folder was not generated by multiqc. This could be due to filtering out of modules or samples."
+fi
+
+if [[ -n "\\$par_output_plots" ]] && [[ -d "\\${out_dir}/\\${report_name}_plots" ]]; then
+    mv "\\${out_dir}/\\${report_name}_plots" "\\$par_output_plots"
+elif [[ -n "\\$par_output_plots" ]] && [[ ! -d "\\${out_dir}/\\${report_name}_plots" ]]; then
+    echo "WARNING: Plots could not be saved because plots folder was not generated by multiqc. This could be due to filtering out of modules or samples."
+fi
+VIASHMAIN
+bash "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/biobox/multiqc",
+    "tag" : "v0.4.2"
+  },
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/nextflow.config b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/nextflow.config
new file mode 100644
index 0000000..a008906
--- /dev/null
+++ b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'multiqc'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v0.4.2'
+  description = 'MultiQC aggregates results from bioinformatics analyses across many samples into a single report.\nIt searches a given directory for analysis logs and compiles a HTML report. It\'s a general use tool, perfect for summarising the output from numerous bioinformatics tools.\n'
+  author = 'Dorien Roosen'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+
diff --git a/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/nextflow_schema.json b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/nextflow_schema.json
new file mode 100644
index 0000000..fec92e4
--- /dev/null
+++ b/target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/nextflow_schema.json
@@ -0,0 +1,334 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "multiqc",
+  "description": "MultiQC aggregates results from bioinformatics analyses across many samples into a single report.\nIt searches a given directory for analysis logs and compiles a HTML report. It's a general use tool, perfect for summarising the output from numerous bioinformatics tools.\n",
+  "type": "object",
+  "$defs": {
+    "input": {
+      "title": "Input",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "exists": true,
+          "description": "File paths to be searched for analysis results to be included in the report.\n",
+          "help_text": "Type: `file`, multiple: `True`, required, direction: `input`, example: `[\"data/results\"]`. "
+        }
+      }
+    },
+    "ouput": {
+      "title": "Ouput",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output_report": {
+          "type": "string",
+          "format": "path",
+          "description": "Filepath of the generated report.\n",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output_report.html\"`, direction: `output`, example: `\"multiqc_report.html\"`. ",
+          "default": "$id.$key.output_report.html"
+        },
+        "output_data": {
+          "type": "string",
+          "format": "path",
+          "description": "Output directory for parsed data files",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output_data\"`, direction: `output`, example: `\"multiqc_data\"`. ",
+          "default": "$id.$key.output_data"
+        },
+        "output_plots": {
+          "type": "string",
+          "format": "path",
+          "description": "Output directory for generated plots",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output_plots\"`, direction: `output`, example: `\"multiqc_plots\"`. ",
+          "default": "$id.$key.output_plots"
+        }
+      }
+    },
+    "modules and analyses to run": {
+      "title": "Modules and analyses to run",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "include_modules": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Use only these module",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"fastqc\";\"cutadapt\"]`. "
+        },
+        "exclude_modules": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Do not use only these modules",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"fastqc\";\"cutadapt\"]`. "
+        },
+        "ignore_analysis": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"run_one/*\";\"run_two/*\"]`. "
+        },
+        "ignore_samples": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"sample_1*\";\"sample_3*\"]`. "
+        },
+        "ignore_symlinks": {
+          "type": "boolean",
+          "description": "Ignore symlinked directories and files",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "sample name handling": {
+      "title": "Sample name handling",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "dirs": {
+          "type": "boolean",
+          "description": "Prepend directory to sample names to avoid clashing filenames",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "dirs_depth": {
+          "type": "integer",
+          "description": "Prepend n directories to sample names",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "full_names": {
+          "type": "boolean",
+          "description": "Do not clean the sample names (leave as full file name)",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "fn_as_s_name": {
+          "type": "boolean",
+          "description": "Use the log filename as the sample name",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "replace_names": {
+          "type": "string",
+          "format": "path",
+          "description": "TSV file to rename sample names during report generation",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"replace_names.tsv\"`. "
+        }
+      }
+    },
+    "report customisation": {
+      "title": "Report Customisation",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "title": {
+          "type": "string",
+          "description": "Report title",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "comment": {
+          "type": "string",
+          "description": "Custom comment, will be printed at the top of the report.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "template": {
+          "type": "string",
+          "description": "Report template to use.\n",
+          "help_text": "Type: `string`, multiple: `False`, choices: ``default`, `gathered`, `geo`, `highcharts`, `sections`, `simple``. ",
+          "enum": [
+            "default",
+            "gathered",
+            "geo",
+            "highcharts",
+            "sections",
+            "simple"
+          ]
+        },
+        "sample_names": {
+          "type": "string",
+          "format": "path",
+          "description": "TSV file containing alternative sample names for renaming buttons in the report.\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"sample_names.tsv\"`. "
+        },
+        "sample_filters": {
+          "type": "string",
+          "format": "path",
+          "description": "TSV file containing show/hide patterns for the report\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"sample_filters.tsv\"`. "
+        },
+        "custom_css_file": {
+          "type": "string",
+          "format": "path",
+          "description": "Custom CSS file to add to the final report\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"custom_style_sheet.css\"`. "
+        },
+        "profile_runtime": {
+          "type": "boolean",
+          "description": "Add analysis of how long MultiQC takes to run to the report\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "multiqc behaviour": {
+      "title": "MultiQC behaviour",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "verbose": {
+          "type": "boolean",
+          "description": "Increase output verbosity.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "quiet": {
+          "type": "boolean",
+          "description": "Only show log warnings\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "strict": {
+          "type": "boolean",
+          "description": "Don't catch exceptions, run additional code checks to help development.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "development": {
+          "type": "boolean",
+          "description": "Development mode",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "require_logs": {
+          "type": "boolean",
+          "description": "Require all explicitly requested modules to have log files",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "no_megaqc_upload": {
+          "type": "boolean",
+          "description": "Don't upload generated report to MegaQC, even if MegaQC options are found.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "no_ansi": {
+          "type": "boolean",
+          "description": "Disable coloured log output.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "cl_config": {
+          "type": "string",
+          "description": "YAML formatted string that allows to customize MultiQC behaviour like input file detection.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"qualimap_config: { general_stats_coverage: [20,40,200] }\"`. "
+        }
+      }
+    },
+    "output format": {
+      "title": "Output format",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "flat": {
+          "type": "boolean",
+          "description": "Use only flat plots (static images).\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "interactive": {
+          "type": "boolean",
+          "description": "Use only interactive plots (in-browser Javascript).\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "data_dir": {
+          "type": "boolean",
+          "description": "Force the parsed data directory to be created.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "no_data_dir": {
+          "type": "boolean",
+          "description": "Prevent the parsed data directory from being created.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "zip_data_dir": {
+          "type": "boolean",
+          "description": "Compress the data directory.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "data_format": {
+          "type": "string",
+          "description": "Output parsed data in a different format than the default 'txt'.\n",
+          "help_text": "Type: `string`, multiple: `False`, choices: ``tsv`, `csv`, `json`, `yaml``. ",
+          "enum": [
+            "tsv",
+            "csv",
+            "json",
+            "yaml"
+          ]
+        },
+        "pdf": {
+          "type": "boolean",
+          "description": "Creates PDF report with the 'simple' template",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/input"
+    },
+    {
+      "$ref": "#/$defs/ouput"
+    },
+    {
+      "$ref": "#/$defs/modules and analyses to run"
+    },
+    {
+      "$ref": "#/$defs/sample name handling"
+    },
+    {
+      "$ref": "#/$defs/report customisation"
+    },
+    {
+      "$ref": "#/$defs/multiqc behaviour"
+    },
+    {
+      "$ref": "#/$defs/output format"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/.config.vsh.yaml b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/.config.vsh.yaml
new file mode 100644
index 0000000..8a58ac6
--- /dev/null
+++ b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/.config.vsh.yaml
@@ -0,0 +1,187 @@
+name: "move_files_to_directory"
+version: "v0.2.0"
+authors:
+- name: "Dorien Roosen"
+  roles:
+  - "maintainer"
+  info:
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Paths of the files that will be copied into the output directory."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    description: "Path to output directory"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "bash_script"
+  path: "script.sh"
+  is_executable: true
+summary: "Publish one or multiple files to the same directory"
+description: "This component copies one or multiple files to the same destination\
+  \ directory, creating the output directory if it doesn't exist."
+test_resources:
+- type: "bash_script"
+  path: "test.sh"
+  is_executable: true
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+license: "MIT"
+links:
+  repository: "https://github.com/viash-hub/craftbox"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "debian:latest"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v0.2.0"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/move_files_to_directory/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/move_files_to_directory"
+  executable: "target/nextflow/move_files_to_directory/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "1c1b0a4a1aff891ab678072b0ba915ac3ac71610"
+  git_remote: "https://github.com/viash-hub/craftbox"
+  git_tag: "v0.1.0-8-g1c1b0a4"
+package_config:
+  name: "craftbox"
+  version: "v0.2.0"
+  summary: "A collection of custom-tailored scripts and applied utilities built with\
+    \ Viash.\n"
+  description: "`craftbox` is a curated collection of custom scripts and utilities\
+    \ designed to tackle context-specific tasks.\n\nEmphasizing the Viash principles,\
+    \ `craftbox` components aim for **reusability**, **reproducibility**, and adherence\
+    \ to **best practices**. Key features generally include:\n\n* **Standalone & Nextflow\
+    \ Ready:** Components are built to run directly via the command line or be smoothly\
+    \ integrated into Nextflow workflows.\n* **Custom Implementations:** Contains\
+    \ scripts and tools developed for particular tasks that may not be found in broader\
+    \ collections.\n* **High Quality Standards (promoted by Viash):**\n    * Clear\
+    \ documentation for components and their parameters.\n    * Full exposure of underlying\
+    \ script/tool arguments for fine-grained control.\n    * Containerized (Docker)\
+    \ to ensure dependency management and a consistent, reproducible runtime environment.\n\
+    \    * Unit tested where applicable to ensure components function as expected.\n"
+  info: null
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v0.2.0'"
+  keywords:
+  - "scripts"
+  - "custom"
+  - "implementations"
+  - "utilities"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/viash-hub/craftbox"
+    issue_tracker: "https://github.com/viash-hub/craftbox/issues"
diff --git a/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/main.nf b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/main.nf
new file mode 100644
index 0000000..f4e3ca1
--- /dev/null
+++ b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/main.nf
@@ -0,0 +1,3811 @@
+// move_files_to_directory v0.2.0
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dorien Roosen (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "move_files_to_directory",
+  "version" : "v0.2.0",
+  "authors" : [
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Paths of the files that will be copied into the output directory.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Path to output directory",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "bash_script",
+      "path" : "script.sh",
+      "is_executable" : true
+    }
+  ],
+  "summary" : "Publish one or multiple files to the same directory",
+  "description" : "This component copies one or multiple files to the same destination directory, creating the output directory if it doesn't exist.",
+  "test_resources" : [
+    {
+      "type" : "bash_script",
+      "path" : "test.sh",
+      "is_executable" : true
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/viash-hub/craftbox"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        }
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "debian:latest",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v0.2.0",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/move_files_to_directory/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "target/nextflow/move_files_to_directory",
+    "viash_version" : "0.9.4",
+    "git_commit" : "1c1b0a4a1aff891ab678072b0ba915ac3ac71610",
+    "git_remote" : "https://github.com/viash-hub/craftbox",
+    "git_tag" : "v0.1.0-8-g1c1b0a4"
+  },
+  "package_config" : {
+    "name" : "craftbox",
+    "version" : "v0.2.0",
+    "summary" : "A collection of custom-tailored scripts and applied utilities built with Viash.\n",
+    "description" : "`craftbox` is a curated collection of custom scripts and utilities designed to tackle context-specific tasks.\n\nEmphasizing the Viash principles, `craftbox` components aim for **reusability**, **reproducibility**, and adherence to **best practices**. Key features generally include:\n\n* **Standalone & Nextflow Ready:** Components are built to run directly via the command line or be smoothly integrated into Nextflow workflows.\n* **Custom Implementations:** Contains scripts and tools developed for particular tasks that may not be found in broader collections.\n* **High Quality Standards (promoted by Viash):**\n    * Clear documentation for components and their parameters.\n    * Full exposure of underlying script/tool arguments for fine-grained control.\n    * Containerized (Docker) to ensure dependency management and a consistent, reproducible runtime environment.\n    * Unit tested where applicable to ensure components function as expected.\n",
+    "viash_version" : "0.9.4",
+    "source" : "src",
+    "target" : "target",
+    "config_mods" : [
+      ".requirements.commands := ['ps']\n",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v0.2.0'"
+    ],
+    "keywords" : [
+      "scripts",
+      "custom",
+      "implementations",
+      "utilities"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/viash-hub/craftbox",
+      "issue_tracker" : "https://github.com/viash-hub/craftbox/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.sh"
+cat > "$tempscript" << VIASHMAIN
+#!/bin/bash
+
+set -eo pipefail
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+$( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "${VIASH_PAR_INPUT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_input='&'#" ; else echo "# par_input="; fi )
+$( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "${VIASH_PAR_OUTPUT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_output='&'#" ; else echo "# par_output="; fi )
+$( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "${VIASH_META_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_name='&'#" ; else echo "# meta_name="; fi )
+$( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "${VIASH_META_FUNCTIONALITY_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_functionality_name='&'#" ; else echo "# meta_functionality_name="; fi )
+$( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "${VIASH_META_RESOURCES_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_resources_dir='&'#" ; else echo "# meta_resources_dir="; fi )
+$( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "${VIASH_META_EXECUTABLE}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_executable='&'#" ; else echo "# meta_executable="; fi )
+$( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "${VIASH_META_CONFIG}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_config='&'#" ; else echo "# meta_config="; fi )
+$( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "${VIASH_META_TEMP_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_temp_dir='&'#" ; else echo "# meta_temp_dir="; fi )
+$( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "${VIASH_META_CPUS}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_cpus='&'#" ; else echo "# meta_cpus="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "${VIASH_META_MEMORY_B}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_b='&'#" ; else echo "# meta_memory_b="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "${VIASH_META_MEMORY_KB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_kb='&'#" ; else echo "# meta_memory_kb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "${VIASH_META_MEMORY_MB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_mb='&'#" ; else echo "# meta_memory_mb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "${VIASH_META_MEMORY_GB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_gb='&'#" ; else echo "# meta_memory_gb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "${VIASH_META_MEMORY_TB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_tb='&'#" ; else echo "# meta_memory_tb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "${VIASH_META_MEMORY_PB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_pb='&'#" ; else echo "# meta_memory_pb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "${VIASH_META_MEMORY_KIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_kib='&'#" ; else echo "# meta_memory_kib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "${VIASH_META_MEMORY_MIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_mib='&'#" ; else echo "# meta_memory_mib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "${VIASH_META_MEMORY_GIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_gib='&'#" ; else echo "# meta_memory_gib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "${VIASH_META_MEMORY_TIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_tib='&'#" ; else echo "# meta_memory_tib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "${VIASH_META_MEMORY_PIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_pib='&'#" ; else echo "# meta_memory_pib="; fi )
+
+## VIASH END
+
+
+if [[ ! -d "\\$par_output" ]]; then
+  mkdir -p "\\$par_output"
+fi
+
+# Process multiple input files
+IFS=";" read -ra input_files <<< "\\$par_input"
+for file in "\\${input_files[@]}"; do
+  # Check if the file exists before copying
+  if [[ -f "\\$file" ]]; then
+    cp "\\$file" "\\$par_output/"
+    echo "Copied \\$file to \\$par_output/"
+  else
+    echo "Warning: Input file \\$file does not exist, skipping"
+  fi
+done
+VIASHMAIN
+bash "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/craftbox/move_files_to_directory",
+    "tag" : "v0.2.0"
+  },
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/nextflow.config b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/nextflow.config
new file mode 100644
index 0000000..38bce9a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'move_files_to_directory'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v0.2.0'
+  description = 'This component copies one or multiple files to the same destination directory, creating the output directory if it doesn\'t exist.'
+  author = 'Dorien Roosen'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+
diff --git a/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/nextflow_schema.json b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/nextflow_schema.json
new file mode 100644
index 0000000..ddc6758
--- /dev/null
+++ b/target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/nextflow_schema.json
@@ -0,0 +1,81 @@
+{
+"$schema": "http://json-schema.org/draft-07/schema",
+"title": "move_files_to_directory",
+"description": "This component copies one or multiple files to the same destination directory, creating the output directory if it doesn\u0027t exist.",
+"type": "object",
+"definitions": {
+
+    
+    
+    "arguments" : {
+    "title": "Arguments",
+    "type": "object",
+    "description": "No description",
+    "properties": {
+    
+        
+                "input": {
+                "type":
+                "string",
+                "description": "Type: List of `file`, required, multiple_sep: `\";\"`. Paths of the files that will be copied into the output directory",
+                "help_text": "Type: List of `file`, required, multiple_sep: `\";\"`. Paths of the files that will be copied into the output directory."
+            
+            }
+    
+
+        ,
+                "output": {
+                "type":
+                "string",
+                "description": "Type: `file`, required, default: `$id.$key.output`. Path to output directory",
+                "help_text": "Type: `file`, required, default: `$id.$key.output`. Path to output directory"
+            ,
+                "default":"$id.$key.output"
+            }
+    
+
+}
+},
+    
+    
+    "nextflow input-output arguments" : {
+    "title": "Nextflow input-output arguments",
+    "type": "object",
+    "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+    "properties": {
+    
+        
+                "publish_dir": {
+                "type":
+                "string",
+                "description": "Type: `string`, required, example: `output/`. Path to an output directory",
+                "help_text": "Type: `string`, required, example: `output/`. Path to an output directory."
+            
+            }
+    
+
+        ,
+                "param_list": {
+                "type":
+                "string",
+                "description": "Type: `string`, example: `my_params.yaml`. Allows inputting multiple parameter sets to initialise a Nextflow channel",
+                "help_text": "Type: `string`, example: `my_params.yaml`. Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.\n\n* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ [\u0027id\u0027: \u0027foo\u0027, \u0027input\u0027: \u0027foo.txt\u0027], [\u0027id\u0027: \u0027bar\u0027, \u0027input\u0027: \u0027bar.txt\u0027] ]`.\n* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.\n* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {\u0027id\u0027: \u0027foo\u0027, \u0027input\u0027: \u0027foo.txt\u0027}, {\u0027id\u0027: \u0027bar\u0027, \u0027input\u0027: \u0027bar.txt\u0027} ]`.\n* A yaml blob can also be passed directly as a string. Example: `--param_list \"[ {\u0027id\u0027: \u0027foo\u0027, \u0027input\u0027: \u0027foo.txt\u0027}, {\u0027id\u0027: \u0027bar\u0027, \u0027input\u0027: \u0027bar.txt\u0027} ]\"`.\n\nWhen passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.",
+                "hidden": true
+            
+            }
+    
+
+}
+}
+},
+"allOf": [
+
+    {
+    "$ref": "#/definitions/arguments"
+    },
+
+    {
+    "$ref": "#/definitions/nextflow input-output arguments"
+    }
+]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/.config.vsh.yaml
new file mode 100644
index 0000000..563b016
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/.config.vsh.yaml
@@ -0,0 +1,659 @@
+name: "cellbender_remove_background"
+namespace: "correction"
+version: "v4.0.0"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file. Data file on which to run tool. Data must be un-filtered:\
+      \ it should include empty droplets."
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "List of modalities to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Full count matrix as an h5mu file, with background RNA removed.\
+      \ This file contains all the original droplet barcodes."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer_output"
+    description: "Output layer"
+    info: null
+    default:
+    - "cellbender_corrected"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_background_fraction"
+    info: null
+    default:
+    - "cellbender_background_fraction"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_cell_probability"
+    info: null
+    default:
+    - "cellbender_cell_probability"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_cell_size"
+    info: null
+    default:
+    - "cellbender_cell_size"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_droplet_efficiency"
+    description: "Name of the column in the .obs dataframe to store the droplet efficiencies\
+      \ in.\n"
+    info: null
+    default:
+    - "cellbender_droplet_efficiency"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_latent_scale"
+    info: null
+    default:
+    - "cellbender_latent_scale"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_ambient_expression"
+    info: null
+    default:
+    - "cellbender_ambient_expression"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_gene_expression_encoding"
+    info: null
+    default:
+    - "cellbender_gene_expression_encoding"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Arguments"
+  arguments:
+  - type: "boolean"
+    name: "--expected_cells_from_qc"
+    description: "Will use the Cell Ranger QC to determine the estimated number of\
+      \ cells"
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--expected_cells"
+    description: "Number of cells expected in the dataset (a rough estimate within\
+      \ a factor of 2 is sufficient)."
+    info: null
+    example:
+    - 1000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--total_droplets_included"
+    description: "The number of droplets from the rank-ordered UMI plot\nthat will\
+      \ have their cell probabilities inferred as an\noutput. Include the droplets\
+      \ which might contain cells.\nDroplets beyond TOTAL_DROPLETS_INCLUDED should\
+      \ be\n'surely empty' droplets.\n"
+    info: null
+    example:
+    - 25000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--force_cell_umi_prior"
+    description: "Ignore CellBender's heuristic prior estimation, and use this prior\
+      \ for UMI counts in cells."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--force_empty_umi_prior"
+    description: "Ignore CellBender's heuristic prior estimation, and use this prior\
+      \ for UMI counts in empty droplets."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--model"
+    description: "Which model is being used for count data.\n\n* 'naive' subtracts\
+      \ the estimated ambient profile.\n* 'simple' does not model either ambient RNA\
+      \ or random barcode swapping (for debugging purposes -- not recommended).\n\
+      * 'ambient' assumes background RNA is incorporated into droplets.\n* 'swapping'\
+      \ assumes background RNA comes from random barcode swapping (via PCR chimeras).\n\
+      * 'full' uses a combined ambient and swapping model.\n"
+    info: null
+    default:
+    - "full"
+    required: false
+    choices:
+    - "naive"
+    - "simple"
+    - "ambient"
+    - "swapping"
+    - "full"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--epochs"
+    description: "Number of epochs to train."
+    info: null
+    default:
+    - 150
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--low_count_threshold"
+    description: "Droplets with UMI counts below this number are completely \nexcluded\
+      \ from the analysis. This can help identify the correct \nprior for empty droplet\
+      \ counts in the rare case where empty \ncounts are extremely high (over 200).\n"
+    info: null
+    default:
+    - 5
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--z_dim"
+    description: "Dimension of latent variable z.\n"
+    info: null
+    default:
+    - 64
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--z_layers"
+    description: "Dimension of hidden layers in the encoder for z.\n"
+    info: null
+    default:
+    - 512
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "double"
+    name: "--training_fraction"
+    description: "Training detail: the fraction of the data used for training.\nThe\
+      \ rest is never seen by the inference algorithm. Speeds up learning.\n"
+    info: null
+    default:
+    - 0.9
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--empty_drop_training_fraction"
+    description: "Training detail: the fraction of the training data each epoch that\
+      \ \nis drawn (randomly sampled) from surely empty droplets.\n"
+    info: null
+    default:
+    - 0.2
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--ignore_features"
+    description: "Integer indices of features to ignore entirely. In the output\n\
+      count matrix, the counts for these features will be unchanged.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "double"
+    name: "--fpr"
+    description: "Target 'delta' false positive rate in [0, 1). Use 0 for a cohort\n\
+      of samples which will be jointly analyzed for differential expression.\nA false\
+      \ positive is a true signal count that is erroneously removed.\nMore background\
+      \ removal is accompanied by more signal removal at\nhigh values of FPR. You\
+      \ can specify multiple values, which will\ncreate multiple output files.\n"
+    info: null
+    default:
+    - 0.01
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--exclude_feature_types"
+    description: "Feature types to ignore during the analysis. These features will\n\
+      be left unchanged in the output file.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "double"
+    name: "--projected_ambient_count_threshold"
+    description: "Controls how many features are included in the analysis, which\n\
+      can lead to a large speedup. If a feature is expected to have less\nthan PROJECTED_AMBIENT_COUNT_THRESHOLD\
+      \ counts total in all cells\n(summed), then that gene is excluded, and it will\
+      \ be unchanged\nin the output count matrix. For example, \nPROJECTED_AMBIENT_COUNT_THRESHOLD\
+      \ = 0 will include all features\nwhich have even a single count in any empty\
+      \ droplet.\n"
+    info: null
+    default:
+    - 0.1
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--learning_rate"
+    description: "Training detail: lower learning rate for inference.\nA OneCycle\
+      \ learning rate schedule is used, where the\nupper learning rate is ten times\
+      \ this value. (For this\nvalue, probably do not exceed 1e-3).\n"
+    info: null
+    default:
+    - 1.0E-4
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--final_elbo_fail_fraction"
+    description: "Training is considered to have failed if \n(best_test_ELBO - final_test_ELBO)/(best_test_ELBO\
+      \ - initial_test_ELBO) > FINAL_ELBO_FAIL_FRACTION.\nTraining will automatically\
+      \ re-run if --num-training-tries > 1.\nBy default, will not fail training based\
+      \ on final_training_ELBO.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--epoch_elbo_fail_fraction"
+    description: "Training is considered to have failed if \n(previous_epoch_test_ELBO\
+      \ - current_epoch_test_ELBO)/(previous_epoch_test_ELBO - initial_train_ELBO)\
+      \ > EPOCH_ELBO_FAIL_FRACTION.\nTraining will automatically re-run if --num-training-tries\
+      \ > 1.\nBy default, will not fail training based on epoch_training_ELBO.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--num_training_tries"
+    description: "Number of times to attempt to train the model. At each subsequent\
+      \ attempt,\nthe learning rate is multiplied by LEARNING_RATE_RETRY_MULT.\n"
+    info: null
+    default:
+    - 1
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--learning_rate_retry_mult"
+    description: "Learning rate is multiplied by this amount each time a new training\n\
+      attempt is made. (This parameter is only used if training fails based\non EPOCH_ELBO_FAIL_FRACTION\
+      \ or FINAL_ELBO_FAIL_FRACTION and\nNUM_TRAINING_TRIES is > 1.) \n"
+    info: null
+    default:
+    - 0.2
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--posterior_batch_size"
+    description: "Training detail: size of batches when creating the posterior.\n\
+      Reduce this to avoid running out of GPU memory creating the posterior\n(will\
+      \ be slower).\n"
+    info: null
+    default:
+    - 128
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--posterior_regulation"
+    description: "Posterior regularization method. (For experts: not required for\
+      \ normal usage,\nsee documentation). \n\n* PRq is approximate quantile-targeting.\n\
+      * PRmu is approximate mean-targeting aggregated over genes (behavior of v0.2.0).\n\
+      * PRmu_gene is approximate mean-targeting per gene.\n"
+    info: null
+    required: false
+    choices:
+    - "PRq"
+    - "PRmu"
+    - "PRmu_gene"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--alpha"
+    description: "Tunable parameter alpha for the PRq posterior regularization method\n\
+      (not normally used: see documentation).\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--q"
+    description: "Tunable parameter q for the CDF threshold estimation method (not\n\
+      normally used: see documentation).\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--estimator"
+    description: "Output denoised count estimation method. (For experts: not required\n\
+      for normal usage, see documentation).\n"
+    info: null
+    default:
+    - "mckp"
+    required: false
+    choices:
+    - "map"
+    - "mean"
+    - "cdf"
+    - "sample"
+    - "mckp"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--estimator_multiple_cpu"
+    description: "Including the flag --estimator-multiple-cpu will use more than one\n\
+      CPU to compute the MCKP output count estimator in parallel (does nothing\nfor\
+      \ other estimators).\n"
+    info: null
+    direction: "input"
+  - type: "boolean"
+    name: "--constant_learning_rate"
+    description: "Including the flag --constant-learning-rate will use the ClippedAdam\n\
+      optimizer instead of the OneCycleLR learning rate schedule, which is\nthe default.\
+      \ Learning is faster with the OneCycleLR schedule.\nHowever, training can easily\
+      \ be continued from a checkpoint for more\nepochs than the initial command specified\
+      \ when using ClippedAdam. On\nthe other hand, if using the OneCycleLR schedule\
+      \ with 150 epochs\nspecified, it is not possible to pick up from that final\
+      \ checkpoint\nand continue training until 250 epochs.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--debug"
+    description: "Including the flag --debug will log extra messages useful for debugging.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--cuda"
+    description: "Including the flag --cuda will run the inference on a\nGPU.\n"
+    info: null
+    direction: "input"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Eliminating technical artifacts from high-throughput single-cell RNA\
+  \ sequencing data.\n\nThis module removes counts due to ambient RNA molecules and\
+  \ random barcode swapping from (raw) UMI-based scRNA-seq count matrices. \nAt the\
+  \ moment, only the count matrices produced by the CellRanger count pipeline is supported.\
+  \ Support for additional tools and protocols \nwill be added in the future. A quick\
+  \ start tutorial can be found here.\n\nFleming et al. 2022, bioRxiv.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "midcpu"
+    - "midmem"
+    - "gpu"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "nvcr.io/nvidia/cuda:11.8.0-devel-ubuntu22.04"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.0"
+  namespace_separator: "/"
+  setup:
+  - type: "docker"
+    run:
+    - "apt update && DEBIAN_FRONTEND=noninteractive apt install -y make build-essential\
+      \ libssl-dev zlib1g-dev libbz2-dev libreadline-dev libsqlite3-dev wget ca-certificates\
+      \ curl llvm libncurses5-dev xz-utils tk-dev libxml2-dev libxmlsec1-dev libffi-dev\
+      \ liblzma-dev mecab-ipadic-utf8 git \\\n&& curl https://pyenv.run | bash \\\n\
+      && pyenv update \\\n&& pyenv install $PYTHON_VERSION \\\n&& pyenv global $PYTHON_VERSION\
+      \ \\\n&& apt-get clean\n"
+    env:
+    - "PYENV_ROOT=\"/root/.pyenv\""
+    - "PATH=\"$PYENV_ROOT/shims:$PYENV_ROOT/bin:$PATH\""
+    - "PYTHON_VERSION=3.7.16"
+  - type: "python"
+    user: false
+    packages:
+    - "lxml~=4.8.0"
+    - "mudata~=0.2.1"
+    - "cellbender~=0.3.0"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/correction/cellbender_remove_background/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/correction/cellbender_remove_background"
+  executable: "target/nextflow/correction/cellbender_remove_background/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "de02293c9e13198622b988dac952b2c8c70a1e35"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.0"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.0'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/main.nf
new file mode 100644
index 0000000..18a5f2c
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/main.nf
@@ -0,0 +1,4531 @@
+// cellbender_remove_background v4.0.0
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "cellbender_remove_background",
+  "namespace" : "correction",
+  "version" : "v4.0.0",
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file. Data file on which to run tool. Data must be un-filtered: it should include empty droplets.",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "List of modalities to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Full count matrix as an h5mu file, with background RNA removed. This file contains all the original droplet barcodes.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer_output",
+          "description" : "Output layer",
+          "default" : [
+            "cellbender_corrected"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_background_fraction",
+          "default" : [
+            "cellbender_background_fraction"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_cell_probability",
+          "default" : [
+            "cellbender_cell_probability"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_cell_size",
+          "default" : [
+            "cellbender_cell_size"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_droplet_efficiency",
+          "description" : "Name of the column in the .obs dataframe to store the droplet efficiencies in.\n",
+          "default" : [
+            "cellbender_droplet_efficiency"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_latent_scale",
+          "default" : [
+            "cellbender_latent_scale"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_ambient_expression",
+          "default" : [
+            "cellbender_ambient_expression"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_gene_expression_encoding",
+          "default" : [
+            "cellbender_gene_expression_encoding"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--expected_cells_from_qc",
+          "description" : "Will use the Cell Ranger QC to determine the estimated number of cells",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--expected_cells",
+          "description" : "Number of cells expected in the dataset (a rough estimate within a factor of 2 is sufficient).",
+          "example" : [
+            1000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--total_droplets_included",
+          "description" : "The number of droplets from the rank-ordered UMI plot\nthat will have their cell probabilities inferred as an\noutput. Include the droplets which might contain cells.\nDroplets beyond TOTAL_DROPLETS_INCLUDED should be\n'surely empty' droplets.\n",
+          "example" : [
+            25000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--force_cell_umi_prior",
+          "description" : "Ignore CellBender's heuristic prior estimation, and use this prior for UMI counts in cells.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--force_empty_umi_prior",
+          "description" : "Ignore CellBender's heuristic prior estimation, and use this prior for UMI counts in empty droplets.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--model",
+          "description" : "Which model is being used for count data.\n\n* 'naive' subtracts the estimated ambient profile.\n* 'simple' does not model either ambient RNA or random barcode swapping (for debugging purposes -- not recommended).\n* 'ambient' assumes background RNA is incorporated into droplets.\n* 'swapping' assumes background RNA comes from random barcode swapping (via PCR chimeras).\n* 'full' uses a combined ambient and swapping model.\n",
+          "default" : [
+            "full"
+          ],
+          "required" : false,
+          "choices" : [
+            "naive",
+            "simple",
+            "ambient",
+            "swapping",
+            "full"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--epochs",
+          "description" : "Number of epochs to train.",
+          "default" : [
+            150
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--low_count_threshold",
+          "description" : "Droplets with UMI counts below this number are completely \nexcluded from the analysis. This can help identify the correct \nprior for empty droplet counts in the rare case where empty \ncounts are extremely high (over 200).\n",
+          "default" : [
+            5
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--z_dim",
+          "description" : "Dimension of latent variable z.\n",
+          "default" : [
+            64
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--z_layers",
+          "description" : "Dimension of hidden layers in the encoder for z.\n",
+          "default" : [
+            512
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--training_fraction",
+          "description" : "Training detail: the fraction of the data used for training.\nThe rest is never seen by the inference algorithm. Speeds up learning.\n",
+          "default" : [
+            0.9
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--empty_drop_training_fraction",
+          "description" : "Training detail: the fraction of the training data each epoch that \nis drawn (randomly sampled) from surely empty droplets.\n",
+          "default" : [
+            0.2
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--ignore_features",
+          "description" : "Integer indices of features to ignore entirely. In the output\ncount matrix, the counts for these features will be unchanged.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--fpr",
+          "description" : "Target 'delta' false positive rate in [0, 1). Use 0 for a cohort\nof samples which will be jointly analyzed for differential expression.\nA false positive is a true signal count that is erroneously removed.\nMore background removal is accompanied by more signal removal at\nhigh values of FPR. You can specify multiple values, which will\ncreate multiple output files.\n",
+          "default" : [
+            0.01
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--exclude_feature_types",
+          "description" : "Feature types to ignore during the analysis. These features will\nbe left unchanged in the output file.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--projected_ambient_count_threshold",
+          "description" : "Controls how many features are included in the analysis, which\ncan lead to a large speedup. If a feature is expected to have less\nthan PROJECTED_AMBIENT_COUNT_THRESHOLD counts total in all cells\n(summed), then that gene is excluded, and it will be unchanged\nin the output count matrix. For example, \nPROJECTED_AMBIENT_COUNT_THRESHOLD = 0 will include all features\nwhich have even a single count in any empty droplet.\n",
+          "default" : [
+            0.1
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--learning_rate",
+          "description" : "Training detail: lower learning rate for inference.\nA OneCycle learning rate schedule is used, where the\nupper learning rate is ten times this value. (For this\nvalue, probably do not exceed 1e-3).\n",
+          "default" : [
+            1.0E-4
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--final_elbo_fail_fraction",
+          "description" : "Training is considered to have failed if \n(best_test_ELBO - final_test_ELBO)/(best_test_ELBO - initial_test_ELBO) > FINAL_ELBO_FAIL_FRACTION.\nTraining will automatically re-run if --num-training-tries > 1.\nBy default, will not fail training based on final_training_ELBO.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--epoch_elbo_fail_fraction",
+          "description" : "Training is considered to have failed if \n(previous_epoch_test_ELBO - current_epoch_test_ELBO)/(previous_epoch_test_ELBO - initial_train_ELBO) > EPOCH_ELBO_FAIL_FRACTION.\nTraining will automatically re-run if --num-training-tries > 1.\nBy default, will not fail training based on epoch_training_ELBO.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--num_training_tries",
+          "description" : "Number of times to attempt to train the model. At each subsequent attempt,\nthe learning rate is multiplied by LEARNING_RATE_RETRY_MULT.\n",
+          "default" : [
+            1
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--learning_rate_retry_mult",
+          "description" : "Learning rate is multiplied by this amount each time a new training\nattempt is made. (This parameter is only used if training fails based\non EPOCH_ELBO_FAIL_FRACTION or FINAL_ELBO_FAIL_FRACTION and\nNUM_TRAINING_TRIES is > 1.) \n",
+          "default" : [
+            0.2
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--posterior_batch_size",
+          "description" : "Training detail: size of batches when creating the posterior.\nReduce this to avoid running out of GPU memory creating the posterior\n(will be slower).\n",
+          "default" : [
+            128
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--posterior_regulation",
+          "description" : "Posterior regularization method. (For experts: not required for normal usage,\nsee documentation). \n\n* PRq is approximate quantile-targeting.\n* PRmu is approximate mean-targeting aggregated over genes (behavior of v0.2.0).\n* PRmu_gene is approximate mean-targeting per gene.\n",
+          "required" : false,
+          "choices" : [
+            "PRq",
+            "PRmu",
+            "PRmu_gene"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--alpha",
+          "description" : "Tunable parameter alpha for the PRq posterior regularization method\n(not normally used: see documentation).\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--q",
+          "description" : "Tunable parameter q for the CDF threshold estimation method (not\nnormally used: see documentation).\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--estimator",
+          "description" : "Output denoised count estimation method. (For experts: not required\nfor normal usage, see documentation).\n",
+          "default" : [
+            "mckp"
+          ],
+          "required" : false,
+          "choices" : [
+            "map",
+            "mean",
+            "cdf",
+            "sample",
+            "mckp"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--estimator_multiple_cpu",
+          "description" : "Including the flag --estimator-multiple-cpu will use more than one\nCPU to compute the MCKP output count estimator in parallel (does nothing\nfor other estimators).\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--constant_learning_rate",
+          "description" : "Including the flag --constant-learning-rate will use the ClippedAdam\noptimizer instead of the OneCycleLR learning rate schedule, which is\nthe default. Learning is faster with the OneCycleLR schedule.\nHowever, training can easily be continued from a checkpoint for more\nepochs than the initial command specified when using ClippedAdam. On\nthe other hand, if using the OneCycleLR schedule with 150 epochs\nspecified, it is not possible to pick up from that final checkpoint\nand continue training until 250 epochs.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--debug",
+          "description" : "Including the flag --debug will log extra messages useful for debugging.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--cuda",
+          "description" : "Including the flag --cuda will run the inference on a\nGPU.\n",
+          "direction" : "input"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Eliminating technical artifacts from high-throughput single-cell RNA sequencing data.\n\nThis module removes counts due to ambient RNA molecules and random barcode swapping from (raw) UMI-based scRNA-seq count matrices. \nAt the moment, only the count matrices produced by the CellRanger count pipeline is supported. Support for additional tools and protocols \nwill be added in the future. A quick start tutorial can be found here.\n\nFleming et al. 2022, bioRxiv.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "midcpu",
+          "midmem",
+          "gpu"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "nvcr.io/nvidia/cuda:11.8.0-devel-ubuntu22.04",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.0",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "docker",
+          "run" : [
+            "apt update && DEBIAN_FRONTEND=noninteractive apt install -y make build-essential libssl-dev zlib1g-dev libbz2-dev libreadline-dev libsqlite3-dev wget ca-certificates curl llvm libncurses5-dev xz-utils tk-dev libxml2-dev libxmlsec1-dev libffi-dev liblzma-dev mecab-ipadic-utf8 git \\\\\n&& curl https://pyenv.run | bash \\\\\n&& pyenv update \\\\\n&& pyenv install $PYTHON_VERSION \\\\\n&& pyenv global $PYTHON_VERSION \\\\\n&& apt-get clean\n"
+          ],
+          "env" : [
+            "PYENV_ROOT=\\"/root/.pyenv\\"",
+            "PATH=\\"$PYENV_ROOT/shims:$PYENV_ROOT/bin:$PATH\\"",
+            "PYTHON_VERSION=3.7.16"
+          ]
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "lxml~=4.8.0",
+            "mudata~=0.2.1",
+            "cellbender~=0.3.0"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/correction/cellbender_remove_background/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/correction/cellbender_remove_background",
+    "viash_version" : "0.9.4",
+    "git_commit" : "de02293c9e13198622b988dac952b2c8c70a1e35",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.0",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.0'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import mudata as mu
+import tempfile
+import subprocess
+import os
+import sys
+import numpy as np
+from scipy.sparse import csr_matrix
+from cellbender.remove_background.downstream import anndata_from_h5
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'layer_output': $( if [ ! -z ${VIASH_PAR_LAYER_OUTPUT+x} ]; then echo "r'${VIASH_PAR_LAYER_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_background_fraction': $( if [ ! -z ${VIASH_PAR_OBS_BACKGROUND_FRACTION+x} ]; then echo "r'${VIASH_PAR_OBS_BACKGROUND_FRACTION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_cell_probability': $( if [ ! -z ${VIASH_PAR_OBS_CELL_PROBABILITY+x} ]; then echo "r'${VIASH_PAR_OBS_CELL_PROBABILITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_cell_size': $( if [ ! -z ${VIASH_PAR_OBS_CELL_SIZE+x} ]; then echo "r'${VIASH_PAR_OBS_CELL_SIZE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_droplet_efficiency': $( if [ ! -z ${VIASH_PAR_OBS_DROPLET_EFFICIENCY+x} ]; then echo "r'${VIASH_PAR_OBS_DROPLET_EFFICIENCY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_latent_scale': $( if [ ! -z ${VIASH_PAR_OBS_LATENT_SCALE+x} ]; then echo "r'${VIASH_PAR_OBS_LATENT_SCALE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_ambient_expression': $( if [ ! -z ${VIASH_PAR_VAR_AMBIENT_EXPRESSION+x} ]; then echo "r'${VIASH_PAR_VAR_AMBIENT_EXPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_gene_expression_encoding': $( if [ ! -z ${VIASH_PAR_OBSM_GENE_EXPRESSION_ENCODING+x} ]; then echo "r'${VIASH_PAR_OBSM_GENE_EXPRESSION_ENCODING//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'expected_cells_from_qc': $( if [ ! -z ${VIASH_PAR_EXPECTED_CELLS_FROM_QC+x} ]; then echo "r'${VIASH_PAR_EXPECTED_CELLS_FROM_QC//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'expected_cells': $( if [ ! -z ${VIASH_PAR_EXPECTED_CELLS+x} ]; then echo "int(r'${VIASH_PAR_EXPECTED_CELLS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'total_droplets_included': $( if [ ! -z ${VIASH_PAR_TOTAL_DROPLETS_INCLUDED+x} ]; then echo "int(r'${VIASH_PAR_TOTAL_DROPLETS_INCLUDED//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'force_cell_umi_prior': $( if [ ! -z ${VIASH_PAR_FORCE_CELL_UMI_PRIOR+x} ]; then echo "int(r'${VIASH_PAR_FORCE_CELL_UMI_PRIOR//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'force_empty_umi_prior': $( if [ ! -z ${VIASH_PAR_FORCE_EMPTY_UMI_PRIOR+x} ]; then echo "int(r'${VIASH_PAR_FORCE_EMPTY_UMI_PRIOR//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'model': $( if [ ! -z ${VIASH_PAR_MODEL+x} ]; then echo "r'${VIASH_PAR_MODEL//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'epochs': $( if [ ! -z ${VIASH_PAR_EPOCHS+x} ]; then echo "int(r'${VIASH_PAR_EPOCHS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'low_count_threshold': $( if [ ! -z ${VIASH_PAR_LOW_COUNT_THRESHOLD+x} ]; then echo "int(r'${VIASH_PAR_LOW_COUNT_THRESHOLD//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'z_dim': $( if [ ! -z ${VIASH_PAR_Z_DIM+x} ]; then echo "int(r'${VIASH_PAR_Z_DIM//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'z_layers': $( if [ ! -z ${VIASH_PAR_Z_LAYERS+x} ]; then echo "list(map(int, r'${VIASH_PAR_Z_LAYERS//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'training_fraction': $( if [ ! -z ${VIASH_PAR_TRAINING_FRACTION+x} ]; then echo "float(r'${VIASH_PAR_TRAINING_FRACTION//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'empty_drop_training_fraction': $( if [ ! -z ${VIASH_PAR_EMPTY_DROP_TRAINING_FRACTION+x} ]; then echo "float(r'${VIASH_PAR_EMPTY_DROP_TRAINING_FRACTION//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'ignore_features': $( if [ ! -z ${VIASH_PAR_IGNORE_FEATURES+x} ]; then echo "list(map(int, r'${VIASH_PAR_IGNORE_FEATURES//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'fpr': $( if [ ! -z ${VIASH_PAR_FPR+x} ]; then echo "list(map(float, r'${VIASH_PAR_FPR//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'exclude_feature_types': $( if [ ! -z ${VIASH_PAR_EXCLUDE_FEATURE_TYPES+x} ]; then echo "r'${VIASH_PAR_EXCLUDE_FEATURE_TYPES//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'projected_ambient_count_threshold': $( if [ ! -z ${VIASH_PAR_PROJECTED_AMBIENT_COUNT_THRESHOLD+x} ]; then echo "float(r'${VIASH_PAR_PROJECTED_AMBIENT_COUNT_THRESHOLD//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'learning_rate': $( if [ ! -z ${VIASH_PAR_LEARNING_RATE+x} ]; then echo "float(r'${VIASH_PAR_LEARNING_RATE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'final_elbo_fail_fraction': $( if [ ! -z ${VIASH_PAR_FINAL_ELBO_FAIL_FRACTION+x} ]; then echo "float(r'${VIASH_PAR_FINAL_ELBO_FAIL_FRACTION//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'epoch_elbo_fail_fraction': $( if [ ! -z ${VIASH_PAR_EPOCH_ELBO_FAIL_FRACTION+x} ]; then echo "float(r'${VIASH_PAR_EPOCH_ELBO_FAIL_FRACTION//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'num_training_tries': $( if [ ! -z ${VIASH_PAR_NUM_TRAINING_TRIES+x} ]; then echo "int(r'${VIASH_PAR_NUM_TRAINING_TRIES//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'learning_rate_retry_mult': $( if [ ! -z ${VIASH_PAR_LEARNING_RATE_RETRY_MULT+x} ]; then echo "float(r'${VIASH_PAR_LEARNING_RATE_RETRY_MULT//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'posterior_batch_size': $( if [ ! -z ${VIASH_PAR_POSTERIOR_BATCH_SIZE+x} ]; then echo "int(r'${VIASH_PAR_POSTERIOR_BATCH_SIZE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'posterior_regulation': $( if [ ! -z ${VIASH_PAR_POSTERIOR_REGULATION+x} ]; then echo "r'${VIASH_PAR_POSTERIOR_REGULATION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'alpha': $( if [ ! -z ${VIASH_PAR_ALPHA+x} ]; then echo "float(r'${VIASH_PAR_ALPHA//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'q': $( if [ ! -z ${VIASH_PAR_Q+x} ]; then echo "float(r'${VIASH_PAR_Q//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'estimator': $( if [ ! -z ${VIASH_PAR_ESTIMATOR+x} ]; then echo "r'${VIASH_PAR_ESTIMATOR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'estimator_multiple_cpu': $( if [ ! -z ${VIASH_PAR_ESTIMATOR_MULTIPLE_CPU+x} ]; then echo "r'${VIASH_PAR_ESTIMATOR_MULTIPLE_CPU//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'constant_learning_rate': $( if [ ! -z ${VIASH_PAR_CONSTANT_LEARNING_RATE+x} ]; then echo "r'${VIASH_PAR_CONSTANT_LEARNING_RATE//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'debug': $( if [ ! -z ${VIASH_PAR_DEBUG+x} ]; then echo "r'${VIASH_PAR_DEBUG//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'cuda': $( if [ ! -z ${VIASH_PAR_CUDA+x} ]; then echo "r'${VIASH_PAR_CUDA//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+
+logger.info("Reading input mudata")
+mdata = mu.read_h5mu(par["input"])
+
+mod = par["modality"]
+logger.info("Performing log transformation on modality %s", mod)
+data = mdata.mod[mod]
+
+# import pathlib
+# with pathlib.Path(os.path.dirname(par["output"])) / "cellbender" as temp_dir:
+#     os.mkdir(temp_dir)
+with tempfile.TemporaryDirectory(
+    prefix="cellbender-", dir=meta["temp_dir"]
+) as temp_dir:
+    # construct paths within tempdir
+    input_file = os.path.join(temp_dir, "input.h5ad")
+    output_file = os.path.join(temp_dir, "output.h5")
+
+    logger.info("Creating AnnData input file for CellBender: '%s'", input_file)
+    data.write_h5ad(input_file)
+
+    logger.info("Constructing CellBender command")
+    cmd_pars = [
+        "cellbender",
+        "remove-background",
+        "--input",
+        input_file,
+        "--output",
+        output_file,
+        # don't create checkpoints because they're not used / returned anyways
+        "--checkpoint-mins",
+        "99999999",
+    ]
+
+    if meta.get("cpus") is not None:
+        cmd_pars += ["--cpu-threads", str(meta["cpus"])]
+
+    extra_args = [
+        ("--expected-cells", "expected_cells", True),
+        ("--total-droplets-included", "total_droplets_included", True),
+        ("--force-cell-umi-prior", "force_cell_umi_prior", True),
+        ("--force-empty-umi-prior", "force_empty_umi_prior", True),
+        ("--model", "model", True),
+        ("--epochs", "epochs", True),
+        ("--low-count-threshold", "low_count_threshold", True),
+        ("--z-dim", "z_dim", True),
+        ("--z-layers", "z_layers", True),
+        ("--training-fraction", "training_fraction", True),
+        ("--empty-drop-training-fraction", "empty_drop_training_fraction", True),
+        ("--ignore-features", "ignore_features", True),
+        ("--fpr", "fpr", True),
+        ("--exclude-feature-types", "exclude_feature_types", True),
+        (
+            "--projected-ambient-count-threshold",
+            "projected_ambient_count_threshold",
+            True,
+        ),
+        ("--learning-rate", "learning_rate", True),
+        ("--final-elbo-fail-fraction", "final_elbo_fail_fraction", True),
+        ("--epoch-elbo-fail-fraction", "epoch_elbo_fail_fraction", True),
+        ("--num-training-tries", "num_training_tries", True),
+        ("--learning-rate-retry-mult", "learning_rate_retry_mult", True),
+        ("--posterior-batch-size", "posterior_batch_size", True),
+        ("--posterior-regulation", "posterior_regulation", True),
+        ("--alpha", "alpha", True),
+        ("--q", "q", True),
+        ("--estimator", "estimator", True),
+        ("--estimator-multiple-cpu", "estimator_multiple_cpu", False),
+        ("--constant-learning-rate", "constant_learning_rate", False),
+        ("--debug", "debug", False),
+        ("--cuda", "cuda", False),
+    ]
+    for flag, name, is_kwarg in extra_args:
+        if par[name]:
+            values = par[name] if isinstance(par[name], list) else [par[name]]
+            cmd_pars += [flag] + [str(val) for val in values] if is_kwarg else [flag]
+
+    if par["expected_cells_from_qc"] and "metrics_cellranger" in data.uns:
+        assert par["expected_cells"] is None, (
+            "If min_counts is defined, expected_cells should be undefined"
+        )
+        assert par["total_droplets_included"] is None, (
+            "If min_counts is defined, expected_cells should be undefined"
+        )
+        met = data.uns["metrics_cellranger"]
+        col_name = "Estimated Number of Cells"
+        assert col_name in met.columns, (
+            "%s should be a column in .obs[metrics_cellranger]"
+        )
+        est_cells = met[col_name].values[0]
+        logger.info(
+            "Selecting --expected-cells %d and --total-droplets-included %d",
+            est_cells,
+            est_cells * 5,
+        )
+        cmd_pars += [
+            "--expected-cells",
+            str(est_cells),
+            "--total-droplets-included",
+            str(5 * est_cells),
+        ]
+
+    logger.info("Running CellBender: '%s'", " ".join(cmd_pars))
+    out = subprocess.check_output(cmd_pars).decode("utf-8")
+
+    logger.info("Reading CellBender 10xh5 output file: '%s'", output_file)
+    adata_out = anndata_from_h5(output_file, analyzed_barcodes_only=False)
+
+    logger.info("CellBender output format:", adata_out)
+
+    # AnnData object with n_obs x n_vars = 6794880 x 33538
+    #     obs: 'cellbender_analyzed'
+    #     var: 'ambient_expression', 'feature_type', 'genome', 'gene_id', 'cellbender_analyzed'
+    #     uns: 'background_fraction', 'barcode_indices_for_latents', 'cell_probability', 'cell_size', 'droplet_efficiency', 'gene_expression_encoding',
+    #          'cell_size_lognormal_std', 'empty_droplet_size_lognormal_loc', 'empty_droplet_size_lognormal_scale', 'swapping_fraction_dist_params',
+    #          'barcodes_analyzed', 'barcodes_analyzed_inds', 'estimator', 'features_analyzed_inds', 'fraction_data_used_for_testing', 'learning_curve_learning_rate_epoch',
+    #          'learning_curve_learning_rate_value', 'learning_curve_test_elbo', 'learning_curve_test_epoch', 'learning_curve_train_elbo', 'learning_curve_train_epoch',
+    #          'target_false_positive_rate'
+
+    logger.info("Copying X output to MuData")
+    data.layers[par["layer_output"]] = adata_out.X
+
+    logger.info("Copying .obs output to MuData")
+    obs_store = {
+        "obs_background_fraction": "background_fraction",
+        "obs_cell_probability": "cell_probability",
+        "obs_cell_size": "cell_size",
+        "obs_droplet_efficiency": "droplet_efficiency",
+        "obs_latent_scale": "latent_scale",
+    }
+    for to_name, from_name in obs_store.items():
+        if par[to_name]:
+            if from_name in adata_out.obs:
+                data.obs[par[to_name]] = adata_out.obs[from_name]
+            # when using unfiltered data, the values will be in uns instead of obs
+            elif (
+                from_name in adata_out.uns
+                and "barcode_indices_for_latents" in adata_out.uns
+            ):
+                vec = np.zeros(data.n_obs)
+                vec[adata_out.uns["barcode_indices_for_latents"]] = adata_out.uns[
+                    from_name
+                ]
+                data.obs[par[to_name]] = vec
+
+    logger.info("Copying .var output to MuData")
+    var_store = {"var_ambient_expression": "ambient_expression"}
+    for to_name, from_name in var_store.items():
+        if par[to_name]:
+            data.var[par[to_name]] = adata_out.var[from_name]
+
+    logger.info("Copying obsm_gene_expression_encoding output to MuData")
+    obsm_store = {"obsm_gene_expression_encoding": "gene_expression_encoding"}
+    for to_name, from_name in obsm_store.items():
+        if par[to_name]:
+            if from_name in adata_out.obsm:
+                data.obsm[par[to_name]] = adata_out.obsm[from_name]
+            elif (
+                from_name in adata_out.uns
+                and "barcode_indices_for_latents" in adata_out.uns
+            ):
+                matrix_to_store = adata_out.uns[from_name]
+                number_of_obs = data.X.shape[0]
+                latent_space_sparse = csr_matrix(
+                    (number_of_obs, par["z_dim"]), dtype=adata_out.uns[from_name].dtype
+                )
+                obs_rows_in_space_representation = adata_out.uns[
+                    "barcode_indices_for_latents"
+                ]
+                latent_space_sparse[obs_rows_in_space_representation] = adata_out.uns[
+                    from_name
+                ]
+                data.obsm[par[to_name]] = latent_space_sparse
+            else:
+                raise RuntimeError(
+                    "Requested to save latent gene encoding, but the data is either missing "
+                    "from cellbender output or in an incorrect format."
+                )
+
+
+logger.info("Writing to file %s", par["output"])
+mdata.write_h5mu(filename=par["output"], compression=par["output_compression"])
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/correction/cellbender_remove_background",
+    "tag" : "v4.0.0"
+  },
+  "label" : [
+    "midcpu",
+    "midmem",
+    "gpu"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/nextflow.config
new file mode 100644
index 0000000..1362212
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/nextflow.config
@@ -0,0 +1,125 @@
+manifest {
+  name = 'correction/cellbender_remove_background'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.0'
+  description = 'Eliminating technical artifacts from high-throughput single-cell RNA sequencing data.\n\nThis module removes counts due to ambient RNA molecules and random barcode swapping from (raw) UMI-based scRNA-seq count matrices. \nAt the moment, only the count matrices produced by the CellRanger count pipeline is supported. Support for additional tools and protocols \nwill be added in the future. A quick start tutorial can be found here.\n\nFleming et al. 2022, bioRxiv.\n'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/nextflow_schema.json
new file mode 100644
index 0000000..ad11d4a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/nextflow_schema.json
@@ -0,0 +1,335 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "cellbender_remove_background",
+  "description": "Eliminating technical artifacts from high-throughput single-cell RNA sequencing data.\n\nThis module removes counts due to ambient RNA molecules and random barcode swapping from (raw) UMI-based scRNA-seq count matrices. \nAt the moment, only the count matrices produced by the CellRanger count pipeline is supported. Support for additional tools and protocols \nwill be added in the future. A quick start tutorial can be found here.\n\nFleming et al. 2022, bioRxiv.\n",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "List of modalities to process.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Full count matrix as an h5mu file, with background RNA removed",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "layer_output": {
+          "type": "string",
+          "description": "Output layer",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"cellbender_corrected\"`. ",
+          "default": "cellbender_corrected"
+        },
+        "obs_background_fraction": {
+          "type": "string",
+          "description": "",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"cellbender_background_fraction\"`. ",
+          "default": "cellbender_background_fraction"
+        },
+        "obs_cell_probability": {
+          "type": "string",
+          "description": "",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"cellbender_cell_probability\"`. ",
+          "default": "cellbender_cell_probability"
+        },
+        "obs_cell_size": {
+          "type": "string",
+          "description": "",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"cellbender_cell_size\"`. ",
+          "default": "cellbender_cell_size"
+        },
+        "obs_droplet_efficiency": {
+          "type": "string",
+          "description": "Name of the column in the .obs dataframe to store the droplet efficiencies in.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"cellbender_droplet_efficiency\"`. ",
+          "default": "cellbender_droplet_efficiency"
+        },
+        "obs_latent_scale": {
+          "type": "string",
+          "description": "",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"cellbender_latent_scale\"`. ",
+          "default": "cellbender_latent_scale"
+        },
+        "var_ambient_expression": {
+          "type": "string",
+          "description": "",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"cellbender_ambient_expression\"`. ",
+          "default": "cellbender_ambient_expression"
+        },
+        "obsm_gene_expression_encoding": {
+          "type": "string",
+          "description": "",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"cellbender_gene_expression_encoding\"`. ",
+          "default": "cellbender_gene_expression_encoding"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "expected_cells_from_qc": {
+          "type": "boolean",
+          "description": "Will use the Cell Ranger QC to determine the estimated number of cells",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "expected_cells": {
+          "type": "integer",
+          "description": "Number of cells expected in the dataset (a rough estimate within a factor of 2 is sufficient).",
+          "help_text": "Type: `integer`, multiple: `False`, example: `1000`. "
+        },
+        "total_droplets_included": {
+          "type": "integer",
+          "description": "The number of droplets from the rank-ordered UMI plot\nthat will have their cell probabilities inferred as an\noutput",
+          "help_text": "Type: `integer`, multiple: `False`, example: `25000`. "
+        },
+        "force_cell_umi_prior": {
+          "type": "integer",
+          "description": "Ignore CellBender's heuristic prior estimation, and use this prior for UMI counts in cells.",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "force_empty_umi_prior": {
+          "type": "integer",
+          "description": "Ignore CellBender's heuristic prior estimation, and use this prior for UMI counts in empty droplets.",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "model": {
+          "type": "string",
+          "description": "Which model is being used for count data.\n\n* 'naive' subtracts the estimated ambient profile.\n* 'simple' does not model either ambient RNA or random barcode swapping (for debugging purposes -- not recommended).\n* 'ambient' assumes background RNA is incorporated into droplets.\n* 'swapping' assumes background RNA comes from random barcode swapping (via PCR chimeras).\n* 'full' uses a combined ambient and swapping model.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"full\"`, choices: ``naive`, `simple`, `ambient`, `swapping`, `full``. ",
+          "enum": [
+            "naive",
+            "simple",
+            "ambient",
+            "swapping",
+            "full"
+          ],
+          "default": "full"
+        },
+        "epochs": {
+          "type": "integer",
+          "description": "Number of epochs to train.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `150`. ",
+          "default": 150
+        },
+        "low_count_threshold": {
+          "type": "integer",
+          "description": "Droplets with UMI counts below this number are completely \nexcluded from the analysis",
+          "help_text": "Type: `integer`, multiple: `False`, default: `5`. ",
+          "default": 5
+        },
+        "z_dim": {
+          "type": "integer",
+          "description": "Dimension of latent variable z.\n",
+          "help_text": "Type: `integer`, multiple: `False`, default: `64`. ",
+          "default": 64
+        },
+        "z_layers": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Dimension of hidden layers in the encoder for z.\n",
+          "help_text": "Type: `integer`, multiple: `True`, default: `[512]`. ",
+          "default": [
+            512
+          ]
+        },
+        "training_fraction": {
+          "type": "number",
+          "description": "Training detail: the fraction of the data used for training.\nThe rest is never seen by the inference algorithm",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.9`. ",
+          "default": 0.9
+        },
+        "empty_drop_training_fraction": {
+          "type": "number",
+          "description": "Training detail: the fraction of the training data each epoch that \nis drawn (randomly sampled) from surely empty droplets.\n",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.2`. ",
+          "default": 0.2
+        },
+        "ignore_features": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Integer indices of features to ignore entirely",
+          "help_text": "Type: `integer`, multiple: `True`. "
+        },
+        "fpr": {
+          "type": "array",
+          "items": {
+            "type": "number"
+          },
+          "description": "Target 'delta' false positive rate in [0, 1)",
+          "help_text": "Type: `double`, multiple: `True`, default: `[0.01]`. ",
+          "default": [
+            0.01
+          ]
+        },
+        "exclude_feature_types": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Feature types to ignore during the analysis",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "projected_ambient_count_threshold": {
+          "type": "number",
+          "description": "Controls how many features are included in the analysis, which\ncan lead to a large speedup",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.1`. ",
+          "default": 0.1
+        },
+        "learning_rate": {
+          "type": "number",
+          "description": "Training detail: lower learning rate for inference.\nA OneCycle learning rate schedule is used, where the\nupper learning rate is ten times this value",
+          "help_text": "Type: `double`, multiple: `False`, default: `1.0E-4`. ",
+          "default": 0.00010
+        },
+        "final_elbo_fail_fraction": {
+          "type": "number",
+          "description": "Training is considered to have failed if \n(best_test_ELBO - final_test_ELBO)/(best_test_ELBO - initial_test_ELBO) > FINAL_ELBO_FAIL_FRACTION.\nTraining will automatically re-run if --num-training-tries > 1.\nBy default, will not fail training based on final_training_ELBO.\n",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "epoch_elbo_fail_fraction": {
+          "type": "number",
+          "description": "Training is considered to have failed if \n(previous_epoch_test_ELBO - current_epoch_test_ELBO)/(previous_epoch_test_ELBO - initial_train_ELBO) > EPOCH_ELBO_FAIL_FRACTION.\nTraining will automatically re-run if --num-training-tries > 1.\nBy default, will not fail training based on epoch_training_ELBO.\n",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "num_training_tries": {
+          "type": "integer",
+          "description": "Number of times to attempt to train the model",
+          "help_text": "Type: `integer`, multiple: `False`, default: `1`. ",
+          "default": 1
+        },
+        "learning_rate_retry_mult": {
+          "type": "number",
+          "description": "Learning rate is multiplied by this amount each time a new training\nattempt is made",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.2`. ",
+          "default": 0.2
+        },
+        "posterior_batch_size": {
+          "type": "integer",
+          "description": "Training detail: size of batches when creating the posterior.\nReduce this to avoid running out of GPU memory creating the posterior\n(will be slower).\n",
+          "help_text": "Type: `integer`, multiple: `False`, default: `128`. ",
+          "default": 128
+        },
+        "posterior_regulation": {
+          "type": "string",
+          "description": "Posterior regularization method",
+          "help_text": "Type: `string`, multiple: `False`, choices: ``PRq`, `PRmu`, `PRmu_gene``. ",
+          "enum": [
+            "PRq",
+            "PRmu",
+            "PRmu_gene"
+          ]
+        },
+        "alpha": {
+          "type": "number",
+          "description": "Tunable parameter alpha for the PRq posterior regularization method\n(not normally used: see documentation).\n",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "q": {
+          "type": "number",
+          "description": "Tunable parameter q for the CDF threshold estimation method (not\nnormally used: see documentation).\n",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "estimator": {
+          "type": "string",
+          "description": "Output denoised count estimation method",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"mckp\"`, choices: ``map`, `mean`, `cdf`, `sample`, `mckp``. ",
+          "enum": [
+            "map",
+            "mean",
+            "cdf",
+            "sample",
+            "mckp"
+          ],
+          "default": "mckp"
+        },
+        "estimator_multiple_cpu": {
+          "type": "boolean",
+          "description": "Including the flag --estimator-multiple-cpu will use more than one\nCPU to compute the MCKP output count estimator in parallel (does nothing\nfor other estimators).\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "constant_learning_rate": {
+          "type": "boolean",
+          "description": "Including the flag --constant-learning-rate will use the ClippedAdam\noptimizer instead of the OneCycleLR learning rate schedule, which is\nthe default",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        },
+        "debug": {
+          "type": "boolean",
+          "description": "Including the flag --debug will log extra messages useful for debugging.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "cuda": {
+          "type": "boolean",
+          "description": "Including the flag --cuda will run the inference on a\nGPU.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/.config.vsh.yaml
new file mode 100644
index 0000000..5de843a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/.config.vsh.yaml
@@ -0,0 +1,274 @@
+name: "add_id"
+namespace: "metadata"
+version: "v4.0.0"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to the input .h5mu."
+    info: null
+    example:
+    - "sample_path"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_id"
+    description: "The input id."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_output"
+    description: "Name of the .obs column where to store the id."
+    info: null
+    default:
+    - "sample_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Name of output MuData file.\n"
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--make_observation_keys_unique"
+    description: "Join the id to the .obs index (.obs_names)."
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Add id of .obs. Also allows to make .obs_names (the .obs index) unique\
+  \ \nby prefixing the values with an unique id per .h5mu file.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "lowmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.11-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.0"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/metadata/add_id/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/metadata/add_id"
+  executable: "target/nextflow/metadata/add_id/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "de02293c9e13198622b988dac952b2c8c70a1e35"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.0"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.0'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/main.nf
new file mode 100644
index 0000000..37499a0
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/main.nf
@@ -0,0 +1,3997 @@
+// add_id v4.0.0
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "add_id",
+  "namespace" : "metadata",
+  "version" : "v4.0.0",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to the input .h5mu.",
+          "example" : [
+            "sample_path"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_id",
+          "description" : "The input id.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_output",
+          "description" : "Name of the .obs column where to store the id.",
+          "default" : [
+            "sample_id"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Name of output MuData file.\n",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--make_observation_keys_unique",
+          "description" : "Join the id to the .obs index (.obs_names).",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Add id of .obs. Also allows to make .obs_names (the .obs index) unique \nby prefixing the values with an unique id per .h5mu file.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data/e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "lowmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.11-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.0",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/metadata/add_id/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/metadata/add_id",
+    "viash_version" : "0.9.4",
+    "git_commit" : "de02293c9e13198622b988dac952b2c8c70a1e35",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.0",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.0'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+from __future__ import annotations
+import sys
+from mudata import read_h5mu, MuData
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_id': $( if [ ! -z ${VIASH_PAR_INPUT_ID+x} ]; then echo "r'${VIASH_PAR_INPUT_ID//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_output': $( if [ ! -z ${VIASH_PAR_OBS_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OBS_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'make_observation_keys_unique': $( if [ ! -z ${VIASH_PAR_MAKE_OBSERVATION_KEYS_UNIQUE+x} ]; then echo "r'${VIASH_PAR_MAKE_OBSERVATION_KEYS_UNIQUE//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+
+def make_observation_keys_unique(sample_id: str, sample: MuData) -> None:
+    """
+    Make the observation keys unique across all samples. At input,
+    the observation keys are unique within a sample. By adding the sample name
+    (unique for a sample) to each observation key, the observation key is made
+    unique across all samples as well.
+    """
+    logger.info(
+        "Making observation keys unique across all "
+        "samples by appending prefix '%s' to the observation names.",
+        sample_id,
+    )
+    sample.obs.index = f"{sample_id}_" + sample.obs.index
+    make_observation_keys_unique_per_mod(sample_id, sample)
+    logger.info("Done making observation keys unique.")
+
+
+def make_observation_keys_unique_per_mod(sample_id: str, sample: MuData) -> None:
+    """
+    Updating MuData.obs_names is not allowed (it is read-only).
+    So the observation keys for each modality has to be updated manually.
+    """
+    for mod_name, mod in sample.mod.items():
+        logger.info("Processing modality '%s'", mod_name)
+        mod.obs_names = f"{sample_id}_" + mod.obs_names
+
+
+def main():
+    logger.info("Reading input file '%s'.", par["input"])
+    input_data = read_h5mu(par["input"])
+    logger.info(
+        "Adding column '%s' to global .obs dataframe, populated with ID '%s'",
+        par["obs_output"],
+        par["input_id"],
+    )
+    input_data.obs[par["obs_output"]] = par["input_id"]
+    logger.info("Done adding column to global .obs")
+    for mod_name, mod_data in input_data.mod.items():
+        logger.info(
+            "Adding column '%s' to .obs dataframe for modality '%s', "
+            "populated with ID '%s'",
+            par["obs_output"],
+            mod_name,
+            par["input_id"],
+        )
+        mod_data.obs[par["obs_output"]] = par["input_id"]
+    logger.info("Done adding per-modality columns.")
+    if par["make_observation_keys_unique"]:
+        make_observation_keys_unique(par["input_id"], input_data)
+    logger.info(
+        "Writing out data to '%s' with compression '%s'.",
+        par["output"],
+        par["output_compression"],
+    )
+    input_data.write_h5mu(par["output"], compression=par["output_compression"])
+    logger.info("Finished")
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/metadata/add_id",
+    "tag" : "v4.0.0"
+  },
+  "label" : [
+    "singlecpu",
+    "lowmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/nextflow.config
new file mode 100644
index 0000000..01e0efb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'metadata/add_id'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.0'
+  description = 'Add id of .obs. Also allows to make .obs_names (the .obs index) unique \nby prefixing the values with an unique id per .h5mu file.\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/nextflow_schema.json
new file mode 100644
index 0000000..83fcf94
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/nextflow_schema.json
@@ -0,0 +1,75 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "add_id",
+  "description": "Add id of .obs. Also allows to make .obs_names (the .obs index) unique \nby prefixing the values with an unique id per .h5mu file.\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the input .h5mu.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"sample_path\"`. "
+        },
+        "input_id": {
+          "type": "string",
+          "description": "The input id.",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        },
+        "obs_output": {
+          "type": "string",
+          "description": "Name of the .obs column where to store the id.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"sample_id\"`. ",
+          "default": "sample_id"
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Name of output MuData file.\n",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "make_observation_keys_unique": {
+          "type": "boolean",
+          "description": "Join the id to the .obs index (.obs_names).",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/.config.vsh.yaml
new file mode 100644
index 0000000..9ab246a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/.config.vsh.yaml
@@ -0,0 +1,330 @@
+name: "grep_annotation_column"
+namespace: "metadata"
+version: "v4.0.0"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  description: "Arguments related to the input dataset."
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to the input .h5mu."
+    info: null
+    example:
+    - "sample_path"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_column"
+    description: "Column to query. If not specified, use .var_names or .obs_names,\
+      \ depending on the value of --matrix"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_layer"
+    description: "Input data to use when calculating fraction of observations that\
+      \ match with the query. \nOnly used when --output_fraction_column is provided.\
+      \ If not specified, .X is used.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality to get the annotation matrix from.\n"
+    info: null
+    example:
+    - "rna"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--matrix"
+    description: "Matrix to fetch the column from that will be searched."
+    info: null
+    example:
+    - "var"
+    required: false
+    choices:
+    - "var"
+    - "obs"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  description: "Arguments related to how the output will be written."
+  arguments:
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Location of the output MuData file.\n"
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_match_column"
+    description: "Name of the column to write the result to."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_fraction_column"
+    description: "For the opposite axis, name of the column to write the fraction\
+      \ of \nobservations that matches to the pattern.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Query options"
+  description: "Options related to the query"
+  arguments:
+  - type: "string"
+    name: "--regex_pattern"
+    description: "Regex to use to match with the input column."
+    info: null
+    example:
+    - "^[mM][tT]-"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Perform a regex lookup on a column from the annotation matrices .obs\
+  \ or .var.\nThe annotation matrix can originate from either a modality, or all modalities\
+  \ (global .var or .obs).\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "lowmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.11-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.0"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/metadata/grep_annotation_column/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/metadata/grep_annotation_column"
+  executable: "target/nextflow/metadata/grep_annotation_column/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "de02293c9e13198622b988dac952b2c8c70a1e35"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.0"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.0'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/main.nf
new file mode 100644
index 0000000..665edca
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/main.nf
@@ -0,0 +1,4136 @@
+// grep_annotation_column v4.0.0
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "grep_annotation_column",
+  "namespace" : "metadata",
+  "version" : "v4.0.0",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "description" : "Arguments related to the input dataset.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to the input .h5mu.",
+          "example" : [
+            "sample_path"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_column",
+          "description" : "Column to query. If not specified, use .var_names or .obs_names, depending on the value of --matrix",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_layer",
+          "description" : "Input data to use when calculating fraction of observations that match with the query. \nOnly used when --output_fraction_column is provided. If not specified, .X is used.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality to get the annotation matrix from.\n",
+          "example" : [
+            "rna"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--matrix",
+          "description" : "Matrix to fetch the column from that will be searched.",
+          "example" : [
+            "var"
+          ],
+          "required" : false,
+          "choices" : [
+            "var",
+            "obs"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "description" : "Arguments related to how the output will be written.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Location of the output MuData file.\n",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_match_column",
+          "description" : "Name of the column to write the result to.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_fraction_column",
+          "description" : "For the opposite axis, name of the column to write the fraction of \nobservations that matches to the pattern.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Query options",
+      "description" : "Options related to the query",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--regex_pattern",
+          "description" : "Regex to use to match with the input column.",
+          "example" : [
+            "^[mM][tT]-"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Perform a regex lookup on a column from the annotation matrices .obs or .var.\nThe annotation matrix can originate from either a modality, or all modalities (global .var or .obs).\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data/e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "lowmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.11-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.0",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/metadata/grep_annotation_column/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/metadata/grep_annotation_column",
+    "viash_version" : "0.9.4",
+    "git_commit" : "de02293c9e13198622b988dac952b2c8c70a1e35",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.0",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.0'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+import mudata as mu
+from pathlib import Path
+from operator import attrgetter
+from pandas import Series
+import scipy as sc
+import re
+import numpy as np
+
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_column': $( if [ ! -z ${VIASH_PAR_INPUT_COLUMN+x} ]; then echo "r'${VIASH_PAR_INPUT_COLUMN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_layer': $( if [ ! -z ${VIASH_PAR_INPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_INPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'matrix': $( if [ ! -z ${VIASH_PAR_MATRIX+x} ]; then echo "r'${VIASH_PAR_MATRIX//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_match_column': $( if [ ! -z ${VIASH_PAR_OUTPUT_MATCH_COLUMN+x} ]; then echo "r'${VIASH_PAR_OUTPUT_MATCH_COLUMN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_fraction_column': $( if [ ! -z ${VIASH_PAR_OUTPUT_FRACTION_COLUMN+x} ]; then echo "r'${VIASH_PAR_OUTPUT_FRACTION_COLUMN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'regex_pattern': $( if [ ! -z ${VIASH_PAR_REGEX_PATTERN+x} ]; then echo "r'${VIASH_PAR_REGEX_PATTERN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+
+def describe_array(arr, msg):
+    # Note: sc.stats returns a DescribeResult NamedTuple. For NamedTuples,
+    # the _asdict method is public facing even though it starts with an underscore.
+    description = sc.stats.describe(arr)._asdict()
+    logger.info(
+        "%s:\\\\nshape: %s\\\\nmean: %s\\\\nnobs: %s\\\\n"
+        "variance: %s\\\\nmin: %s\\\\nmax: %s\\\\ncontains na: %s\\\\ndtype: %s\\\\ncontains 0: %s",
+        msg,
+        arr.shape,
+        description["mean"],
+        description["nobs"],
+        description["variance"],
+        description["minmax"][0],
+        description["minmax"][1],
+        np.isnan(arr).any(),
+        arr.dtype,
+        (arr == 0).any(),
+    )
+
+
+def main(par):
+    input_file, output_file, mod_name = (
+        Path(par["input"]),
+        Path(par["output"]),
+        par["modality"],
+    )
+    logger.info(f"Compiling regular expression '{par['regex_pattern']}'.")
+    try:
+        compiled_regex = re.compile(par["regex_pattern"])
+    except (TypeError, re.error) as e:
+        raise ValueError(
+            f"{par['regex_pattern']} is not a valid regular expression pattern."
+        ) from e
+    else:
+        if compiled_regex.groups:
+            raise NotImplementedError(
+                "Using match groups is not supported by this component."
+            )
+    logger.info("Reading input file %s, modality %s.", input_file, mod_name)
+
+    modality_data = mu.read_h5ad(input_file, mod=mod_name)
+    logger.info("Reading input file done.")
+    logger.info("Using annotation dataframe '%s'.", par["matrix"])
+    annotation_matrix = getattr(modality_data, par["matrix"])
+    default_column = {"var": attrgetter("var_names"), "obs": attrgetter("obs_names")}
+    if par["input_column"]:
+        logger.info("Input column '%s' was specified.", par["input_column"])
+        try:
+            annotation_column = annotation_matrix[par["input_column"]]
+        except KeyError as e:
+            raise ValueError(
+                f"Column {par['input_column']} could not be found for modality "
+                f"{par['modality']}. Available columns:"
+                f" {','.join(annotation_matrix.columns.to_list())}"
+            ) from e
+    else:
+        logger.info(f"No input column specified, using '.{par['matrix']}_names'")
+        annotation_column = default_column[par["matrix"]](modality_data).to_series()
+    logger.info("Applying regex search.")
+    grep_result = annotation_column.str.contains(par["regex_pattern"], regex=True)
+    logger.info("Search results: %s", grep_result.value_counts())
+    # A Series object cannot be used as an indexer for a scipy sparse array
+    # when the data type is a pandas boolean extension array because
+    # extension arrays do not define .nonzero()
+    # See https://github.com/pandas-dev/pandas/issues/46025
+    grep_result = grep_result.to_numpy(dtype="bool", na_value=False)
+
+    other_axis_attribute = {"var": "obs", "obs": "var"}
+    if par["output_fraction_column"]:
+        logger.info(
+            "Enabled writing the fraction of values that matches to the pattern."
+        )
+        input_layer = (
+            modality_data.X
+            if not par["input_layer"]
+            else modality_data.layers[par["input_layer"]]
+        )
+        totals = np.ravel(input_layer.sum(axis=1))
+        describe_array(totals, "Summary of total counts for layer")
+        counts_for_matches = np.ravel(input_layer[:, grep_result].sum(axis=1))
+        describe_array(counts_for_matches, "Summary of counts matching grep")
+        with np.errstate(all="raise"):
+            pct_matching = np.divide(
+                counts_for_matches,
+                totals,
+                out=np.zeros_like(totals, dtype=np.float64),
+                where=(~np.isclose(totals, np.zeros_like(totals))),
+            )
+        logger.info("Testing wether or not fractions data contains NA.")
+        assert ~np.isnan(pct_matching).any(), "Fractions should not contain NA."
+        logger.info("Fraction statistics: \\\\n%s", Series(pct_matching).describe())
+        pct_matching = np.where(np.isclose(pct_matching, 0, atol=1e-6), 0, pct_matching)
+        pct_matching = np.where(np.isclose(pct_matching, 1, atol=1e-6), 1, pct_matching)
+        assert (np.logical_and(pct_matching >= 0, pct_matching <= 1)).all(), (
+            "Fractions are not within bounds, please report this as a bug"
+        )
+        output_matrix = other_axis_attribute[par["matrix"]]
+        logger.info(
+            "Writing fractions to matrix '%s', column '%s'",
+            output_matrix,
+            par["output_fraction_column"],
+        )
+        getattr(modality_data, output_matrix)[par["output_fraction_column"]] = (
+            pct_matching
+        )
+    logger.info(
+        "Adding values that matched the pattern to '%s', column '%s'",
+        par["matrix"],
+        par["output_match_column"],
+    )
+    getattr(modality_data, par["matrix"])[par["output_match_column"]] = grep_result
+    logger.info(
+        "Writing out data to '%s' with compression '%s'.",
+        output_file,
+        par["output_compression"],
+    )
+    write_h5ad_to_h5mu_with_compression(
+        output_file, par["input"], mod_name, modality_data, par["output_compression"]
+    )
+
+
+if __name__ == "__main__":
+    main(par)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/metadata/grep_annotation_column",
+    "tag" : "v4.0.0"
+  },
+  "label" : [
+    "singlecpu",
+    "lowmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/nextflow.config
new file mode 100644
index 0000000..5d333ac
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'metadata/grep_annotation_column'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.0'
+  description = 'Perform a regex lookup on a column from the annotation matrices .obs or .var.\nThe annotation matrix can originate from either a modality, or all modalities (global .var or .obs).\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/nextflow_schema.json
new file mode 100644
index 0000000..ff552cb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/nextflow_schema.json
@@ -0,0 +1,117 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "grep_annotation_column",
+  "description": "Perform a regex lookup on a column from the annotation matrices .obs or .var.\nThe annotation matrix can originate from either a modality, or all modalities (global .var or .obs).\n",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "Arguments related to the input dataset.",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the input .h5mu.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"sample_path\"`. "
+        },
+        "input_column": {
+          "type": "string",
+          "description": "Column to query",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "input_layer": {
+          "type": "string",
+          "description": "Input data to use when calculating fraction of observations that match with the query",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality to get the annotation matrix from.\n",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"rna\"`. "
+        },
+        "matrix": {
+          "type": "string",
+          "description": "Matrix to fetch the column from that will be searched.",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"var\"`, choices: ``var`, `obs``. ",
+          "enum": [
+            "var",
+            "obs"
+          ]
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "Arguments related to how the output will be written.",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Location of the output MuData file.\n",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "output_match_column": {
+          "type": "string",
+          "description": "Name of the column to write the result to.",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        },
+        "output_fraction_column": {
+          "type": "string",
+          "description": "For the opposite axis, name of the column to write the fraction of \nobservations that matches to the pattern.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "query options": {
+      "title": "Query options",
+      "type": "object",
+      "description": "Options related to the query",
+      "properties": {
+        "regex_pattern": {
+          "type": "string",
+          "description": "Regex to use to match with the input column.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"^[mM][tT]-\"`. "
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/query options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/grep_annotation_column/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/.config.vsh.yaml
new file mode 100644
index 0000000..1656546
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/.config.vsh.yaml
@@ -0,0 +1,390 @@
+name: "calculate_qc_metrics"
+namespace: "qc"
+version: "v4.0.0"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process. \n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Layer from modality to use as input data. If not provided the .X\
+      \ attribute is used.\n"
+    info: null
+    example:
+    - "raw_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Metrics added to .obs"
+  arguments:
+  - type: "string"
+    name: "--var_qc_metrics"
+    description: "Keys to select a boolean (containing only True or False) column\
+      \ from .var.\nFor each cell, calculate the proportion of total values for genes\
+      \ which are labeled 'True', \ncompared to the total sum of the values for all\
+      \ genes.\n"
+    info: null
+    example:
+    - "ercc,highly_variable,mitochondrial"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--var_qc_metrics_fill_na_value"
+    description: "Fill any 'NA' values found in the columns specified with --var_qc_metrics\
+      \ to 'True' or 'False'.\nas False.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--top_n_vars"
+    description: "Number of top vars to be used to calculate cumulative proportions.\n\
+      If not specified, proportions are not calculated. `--top_n_vars 20;50` finds\n\
+      cumulative proportion to the 20th and 50th most expressed vars.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_num_nonzero_vars"
+    description: "Name of column in .obs describing, for each observation, the number\
+      \ of stored values\n(including explicit zeroes). In other words, the name of\
+      \ the column that counts\nfor each row the number of columns that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_vars"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_total_counts_vars"
+    description: "Name of the column for .obs describing, for each observation (row),\n\
+      the sum of the stored values in the columns.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Metrics added to .var"
+  arguments:
+  - type: "string"
+    name: "--output_var_num_nonzero_obs"
+    description: "Name of column describing, for each feature, the number of stored\
+      \ values\n(including explicit zeroes). In other words, the name of the column\
+      \ that counts\nfor each column the number of rows that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_obs"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_total_counts_obs"
+    description: "Name of the column in .var describing, for each feature (column),\n\
+      the sum of the stored values in the rows.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_obs_mean"
+    description: "Name of the column in .obs providing the mean of the values in each\
+      \ row.\n"
+    info: null
+    default:
+    - "obs_mean"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_pct_dropout"
+    description: "Name of the column in .obs providing for each feature the percentage\
+      \ of\nobservations the feature does not appear on (i.e. is missing). Same as\
+      \ `--num_nonzero_obs`\nbut percentage based.\n"
+    info: null
+    default:
+    - "pct_dropout"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Output h5mu file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Add basic quality control metrics to an .h5mu file.\n\nThe metrics are\
+  \ comparable to what scanpy.pp.calculate_qc_metrics output,\nalthough they have\
+  \ slightly different names:\n\nVar metrics (name in this component -> name in scanpy):\n\
+  \  - pct_dropout -> pct_dropout_by_{expr_type}\n  - num_nonzero_obs -> n_cells_by_{expr_type}\n\
+  \  - obs_mean -> mean_{expr_type}\n  - total_counts -> total_{expr_type}\n\n  Obs\
+  \ metrics:\n    - num_nonzero_vars -> n_genes_by_{expr_type}\n    - pct_{var_qc_metrics}\
+  \ -> pct_{expr_type}_{qc_var}\n    - total_counts_{var_qc_metrics} -> total_{expr_type}_{qc_var}\n\
+  \    - pct_of_counts_in_top_{top_n_vars}_vars -> pct_{expr_type}_in_top_{n}_{var_type}\n\
+  \    - total_counts -> total_{expr_type}\n    \n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "midmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.11-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.0"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scipy"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  - type: "python"
+    user: false
+    packages:
+    - "scanpy"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/qc/calculate_qc_metrics/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/qc/calculate_qc_metrics"
+  executable: "target/nextflow/qc/calculate_qc_metrics/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "de02293c9e13198622b988dac952b2c8c70a1e35"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.0"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.0'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/main.nf
new file mode 100644
index 0000000..d5e8075
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/main.nf
@@ -0,0 +1,4233 @@
+// calculate_qc_metrics v4.0.0
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "calculate_qc_metrics",
+  "namespace" : "qc",
+  "version" : "v4.0.0",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process. \n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Layer from modality to use as input data. If not provided the .X attribute is used.\n",
+          "example" : [
+            "raw_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Metrics added to .obs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_qc_metrics",
+          "description" : "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes.\n",
+          "example" : [
+            "ercc,highly_variable,mitochondrial"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--var_qc_metrics_fill_na_value",
+          "description" : "Fill any 'NA' values found in the columns specified with --var_qc_metrics to 'True' or 'False'.\nas False.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--top_n_vars",
+          "description" : "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated. `--top_n_vars 20;50` finds\ncumulative proportion to the 20th and 50th most expressed vars.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_num_nonzero_vars",
+          "description" : "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each row the number of columns that contain data.\n",
+          "default" : [
+            "num_nonzero_vars"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_total_counts_vars",
+          "description" : "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Metrics added to .var",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--output_var_num_nonzero_obs",
+          "description" : "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each column the number of rows that contain data.\n",
+          "default" : [
+            "num_nonzero_obs"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_total_counts_obs",
+          "description" : "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_obs_mean",
+          "description" : "Name of the column in .obs providing the mean of the values in each row.\n",
+          "default" : [
+            "obs_mean"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_pct_dropout",
+          "description" : "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e. is missing). Same as `--num_nonzero_obs`\nbut percentage based.\n",
+          "default" : [
+            "pct_dropout"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Output h5mu file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Add basic quality control metrics to an .h5mu file.\n\nThe metrics are comparable to what scanpy.pp.calculate_qc_metrics output,\nalthough they have slightly different names:\n\nVar metrics (name in this component -> name in scanpy):\n  - pct_dropout -> pct_dropout_by_{expr_type}\n  - num_nonzero_obs -> n_cells_by_{expr_type}\n  - obs_mean -> mean_{expr_type}\n  - total_counts -> total_{expr_type}\n\n  Obs metrics:\n    - num_nonzero_vars -> n_genes_by_{expr_type}\n    - pct_{var_qc_metrics} -> pct_{expr_type}_{qc_var}\n    - total_counts_{var_qc_metrics} -> total_{expr_type}_{qc_var}\n    - pct_of_counts_in_top_{top_n_vars}_vars -> pct_{expr_type}_in_top_{n}_{var_type}\n    - total_counts -> total_{expr_type}\n    \n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "midmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.11-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.0",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scipy"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "scanpy"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/qc/calculate_qc_metrics/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/qc/calculate_qc_metrics",
+    "viash_version" : "0.9.4",
+    "git_commit" : "de02293c9e13198622b988dac952b2c8c70a1e35",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.0",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.0'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+from mudata import read_h5ad
+from scipy.sparse import issparse, csr_array
+import numpy as np
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'layer': $( if [ ! -z ${VIASH_PAR_LAYER+x} ]; then echo "r'${VIASH_PAR_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_qc_metrics': $( if [ ! -z ${VIASH_PAR_VAR_QC_METRICS+x} ]; then echo "r'${VIASH_PAR_VAR_QC_METRICS//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'var_qc_metrics_fill_na_value': $( if [ ! -z ${VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE+x} ]; then echo "r'${VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'top_n_vars': $( if [ ! -z ${VIASH_PAR_TOP_N_VARS+x} ]; then echo "list(map(int, r'${VIASH_PAR_TOP_N_VARS//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'output_obs_num_nonzero_vars': $( if [ ! -z ${VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_obs_total_counts_vars': $( if [ ! -z ${VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_var_num_nonzero_obs': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_var_total_counts_obs': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_var_obs_mean': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_OBS_MEAN+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_OBS_MEAN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_var_pct_dropout': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+
+def count_nonzero(layer, axis):
+    """
+    This method is the functional equivalent of the old .getnnz function from scirpy,
+    but that function was deprecated. So we use the nonzero function to mimic the old
+    behavior.
+    """
+    axis ^= 1
+    nonzero_counts = dict(zip(*np.unique(layer.nonzero()[axis], return_counts=True)))
+    nonzero_per_axis_item = {
+        row_index: nonzero_counts.get(row_index, 0)
+        for row_index in range(layer.shape[axis])
+    }
+    return np.array(list(nonzero_per_axis_item.values()), dtype="int64")
+
+
+def main():
+    modality_data = read_h5ad(par["input"], mod=par["modality"])
+    var = modality_data.var
+    layer = modality_data.X if not par["layer"] else modality_data.layers[par["layer"]]
+    if not issparse(layer):
+        raise NotImplementedError("Expected layer to be in sparse format.")
+    layer = csr_array(layer)
+    layer.eliminate_zeros()
+
+    var_columns_to_add = {}
+
+    # var statistics
+    if par["output_var_obs_mean"]:
+        obs_mean = layer.mean(axis=0)
+        var_columns_to_add[par["output_var_obs_mean"]] = obs_mean
+    if par["output_var_total_counts_obs"]:
+        # from the np.sum documentation:
+        # Especially when summing a large number of lower precision floating point numbers,
+        # such as float32, numerical errors can become significant. In such cases it can
+        # be advisable to use dtype="float64" to use a higher precision for the output.
+        layer_with_type = layer
+        if np.issubdtype(layer.dtype, np.floating) and np.can_cast(
+            layer.dtype, np.float64, casting="safe"
+        ):
+            # 'safe' casting makes sure not to cast np.float128 or anything else to a lower precision dtype
+            layer_with_type = layer.astype(np.float64)
+        total_counts_obs = np.ravel(layer_with_type.sum(axis=0))
+        var_columns_to_add[par["output_var_total_counts_obs"]] = total_counts_obs
+
+    num_nonzero_obs = count_nonzero(layer, axis=0)
+    if par["output_var_num_nonzero_obs"]:
+        var_columns_to_add[par["output_var_num_nonzero_obs"]] = num_nonzero_obs
+    if par["output_var_pct_dropout"]:
+        var_columns_to_add[par["output_var_pct_dropout"]] = (
+            1 - num_nonzero_obs / layer.shape[0]
+        ) * 100
+
+    modality_data.var = modality_data.var.assign(**var_columns_to_add)
+
+    # obs statistics
+    obs_columns_to_add = {}
+    total_counts_var = np.ravel(layer.sum(axis=1))
+
+    if par["output_obs_num_nonzero_vars"]:
+        num_nonzero_vars = count_nonzero(layer, axis=1)
+        obs_columns_to_add[par["output_obs_num_nonzero_vars"]] = num_nonzero_vars
+
+    if par["output_obs_total_counts_vars"]:
+        obs_columns_to_add[par["output_obs_total_counts_vars"]] = total_counts_var
+
+    top_metrics = {}
+    if par["top_n_vars"]:
+        par["top_n_vars"] = sorted(par["top_n_vars"])
+        distributions = get_top_from_csr_matrix(layer, par["top_n_vars"])
+        top_metrics = {
+            distribution_size: distribution * 100
+            for distribution_size, distribution in zip(
+                par["top_n_vars"], distributions.T
+            )
+        }
+        obs_columns_to_add |= {
+            f"pct_of_counts_in_top_{n_top}_vars": col
+            for n_top, col in top_metrics.items()
+        }
+
+    if par["var_qc_metrics"]:
+        print(f"qc_metrics: {par['var_qc_metrics']}")
+        for qc_metric in par["var_qc_metrics"]:
+            if qc_metric not in var:
+                raise ValueError(
+                    f"Value for --var_qc_metrics, '{qc_metric}' "
+                    f"not found in .var for modality {par['modality']}"
+                )
+            qc_column = var[qc_metric]
+            if qc_column.isna().any():
+                if par["var_qc_metrics_fill_na_value"] is None:
+                    raise ValueError(
+                        f"The .var column '{qc_metric}', selected by '--var_qc_metrics', contains NA values. "
+                        "It is ambiguous whether or not to include these values in the static calulation. "
+                        "You can explicitly map the NA values to 'False' or 'True using '--var_qc_metrics_fill_na_value'"
+                    )
+                else:
+                    qc_column = qc_column.fillna(
+                        par["var_qc_metrics_fill_na_value"], inplace=False
+                    )
+            qc_column = qc_column.to_list()
+            if set(np.unique(qc_column)) - {True, False}:
+                raise ValueError(
+                    f"Column {qc_metric} in .var for modality {par['modality']} "
+                    f"must only contain boolean values"
+                )
+            total_counts_qc_metric = np.ravel(layer[:, qc_column].sum(axis=1))
+            obs_columns_to_add |= {
+                f"total_counts_{qc_metric}": total_counts_qc_metric,
+                f"pct_{qc_metric}": total_counts_qc_metric / total_counts_var * 100,
+            }
+
+    modality_data.obs = modality_data.obs.assign(**obs_columns_to_add)
+    write_h5ad_to_h5mu_with_compression(
+        par["output"],
+        par["input"],
+        par["modality"],
+        modality_data,
+        par["output_compression"],
+    )
+
+
+def get_top_from_csr_matrix(array, top_n_genes):
+    # csr matrices stores a 3D matrix in a format such that data for individual cells
+    # are stored in 1 array. Another array (indptr) here stores the ranges of indices
+    # to select from the data-array (.e.g. data[indptr[0]:indptr[1]] for row 0) for each row.
+    # Another array 'indices' maps each element of data to a column
+    # (data and indices arrays have the same length)
+    top_n_genes = np.array(top_n_genes).astype(np.int64)
+    assert np.all(top_n_genes[:-1] <= top_n_genes[1:]), "top_n_genes must be sorted"
+    row_indices, data = array.indptr, array.data
+    number_of_rows, max_genes_to_parse = row_indices.size - 1, top_n_genes[-1]
+    top_data = np.zeros((number_of_rows, max_genes_to_parse), dtype=data.dtype)
+    # Loop over each row to create a dense matrix without the 0 counts,
+    # but not for the whole matrix, only store the genes up until
+    # the largest number of top n genes.
+    for row_number in range(number_of_rows):
+        row_start_index, row_end_index = (
+            row_indices[row_number],
+            row_indices[row_number + 1],
+        )
+        row_data = data[row_start_index:row_end_index]  # all non-zero counts for an row
+        try:
+            # There are less genes with counts in the row than the
+            # maximum number of genes we would like to select
+            # all these genes are in the top genes, just store them
+            top_data[row_number, : row_end_index - row_start_index] = row_data
+        except ValueError:
+            # Store the counts for the top genes
+            top_data[row_number, :] = np.partition(row_data, -max_genes_to_parse)[
+                -max_genes_to_parse:
+            ]
+
+    # Partition works from smallest to largest, but we want largest
+    # so do smallest to largest first (but with reversed indices)
+    top_data = np.partition(top_data, max_genes_to_parse - top_n_genes)
+    # And then switch the order around
+    top_data = np.flip(top_data, axis=1)
+
+    cumulative = top_data.cumsum(axis=1, dtype=np.float64)[:, top_n_genes - 1]
+    return cumulative / np.expand_dims(array.sum(axis=1), 1)
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/qc/calculate_qc_metrics",
+    "tag" : "v4.0.0"
+  },
+  "label" : [
+    "singlecpu",
+    "midmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/nextflow.config
new file mode 100644
index 0000000..e002241
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'qc/calculate_qc_metrics'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.0'
+  description = 'Add basic quality control metrics to an .h5mu file.\n\nThe metrics are comparable to what scanpy.pp.calculate_qc_metrics output,\nalthough they have slightly different names:\n\nVar metrics (name in this component -> name in scanpy):\n  - pct_dropout -> pct_dropout_by_{expr_type}\n  - num_nonzero_obs -> n_cells_by_{expr_type}\n  - obs_mean -> mean_{expr_type}\n  - total_counts -> total_{expr_type}\n\n  Obs metrics:\n    - num_nonzero_vars -> n_genes_by_{expr_type}\n    - pct_{var_qc_metrics} -> pct_{expr_type}_{qc_var}\n    - total_counts_{var_qc_metrics} -> total_{expr_type}_{qc_var}\n    - pct_of_counts_in_top_{top_n_vars}_vars -> pct_{expr_type}_in_top_{n}_{var_type}\n    - total_counts -> total_{expr_type}\n    \n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/nextflow_schema.json
new file mode 100644
index 0000000..83f9504
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/nextflow_schema.json
@@ -0,0 +1,156 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "calculate_qc_metrics",
+  "description": "Add basic quality control metrics to an .h5mu file.\n\nThe metrics are comparable to what scanpy.pp.calculate_qc_metrics output,\nalthough they have slightly different names:\n\nVar metrics (name in this component -> name in scanpy):\n  - pct_dropout -> pct_dropout_by_{expr_type}\n  - num_nonzero_obs -> n_cells_by_{expr_type}\n  - obs_mean -> mean_{expr_type}\n  - total_counts -> total_{expr_type}\n\n  Obs metrics:\n    - num_nonzero_vars -> n_genes_by_{expr_type}\n    - pct_{var_qc_metrics} -> pct_{expr_type}_{qc_var}\n    - total_counts_{var_qc_metrics} -> total_{expr_type}_{qc_var}\n    - pct_of_counts_in_top_{top_n_vars}_vars -> pct_{expr_type}_in_top_{n}_{var_type}\n    - total_counts -> total_{expr_type}\n    \n",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "layer": {
+          "type": "string",
+          "description": "Layer from modality to use as input data",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"raw_counts\"`. "
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "metrics added to .obs": {
+      "title": "Metrics added to .obs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_qc_metrics": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes.\n",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"ercc,highly_variable,mitochondrial\"]`. "
+        },
+        "var_qc_metrics_fill_na_value": {
+          "type": "boolean",
+          "description": "Fill any 'NA' values found in the columns specified with --var_qc_metrics to 'True' or 'False'.\nas False.\n",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        },
+        "top_n_vars": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated",
+          "help_text": "Type: `integer`, multiple: `True`. "
+        },
+        "output_obs_num_nonzero_vars": {
+          "type": "string",
+          "description": "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes)",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"num_nonzero_vars\"`. ",
+          "default": "num_nonzero_vars"
+        },
+        "output_obs_total_counts_vars": {
+          "type": "string",
+          "description": "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"total_counts\"`. ",
+          "default": "total_counts"
+        }
+      }
+    },
+    "metrics added to .var": {
+      "title": "Metrics added to .var",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output_var_num_nonzero_obs": {
+          "type": "string",
+          "description": "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes)",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"num_nonzero_obs\"`. ",
+          "default": "num_nonzero_obs"
+        },
+        "output_var_total_counts_obs": {
+          "type": "string",
+          "description": "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"total_counts\"`. ",
+          "default": "total_counts"
+        },
+        "output_var_obs_mean": {
+          "type": "string",
+          "description": "Name of the column in .obs providing the mean of the values in each row.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"obs_mean\"`. ",
+          "default": "obs_mean"
+        },
+        "output_var_pct_dropout": {
+          "type": "string",
+          "description": "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"pct_dropout\"`. ",
+          "default": "pct_dropout"
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/metrics added to .obs"
+    },
+    {
+      "$ref": "#/$defs/metrics added to .var"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/qc/calculate_qc_metrics/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/.config.vsh.yaml
new file mode 100644
index 0000000..f6268a3
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/.config.vsh.yaml
@@ -0,0 +1,415 @@
+name: "qc"
+namespace: "workflows/qc"
+version: "v4.0.0"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the sample."
+    info: null
+    example:
+    - "foo"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to the sample."
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Layer to calculate qc metrics for."
+    info: null
+    example:
+    - "raw_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Mitochondrial & Ribosomal Gene Detection"
+  arguments:
+  - type: "string"
+    name: "--var_gene_names"
+    description: ".var column name to be used to detect mitochondrial/ribosomal genes\
+      \ instead of .var_names (default if not set).\nGene names matching with the\
+      \ regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will\
+      \ be \nidentified as mitochondrial or ribosomal genes, respectively.\n"
+    info: null
+    example:
+    - "gene_symbol"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_mitochondrial_genes"
+    description: "In which .var slot to store a boolean array corresponding the mitochondrial\
+      \ genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_mitochondrial_fraction"
+    description: ".Obs slot to store the fraction of reads found to be mitochondrial.\
+      \ Defaults to 'fraction_' suffixed by the value of --var_name_mitochondrial_genes\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--mitochondrial_gene_regex"
+    description: "Regex string that identifies mitochondrial genes from --var_gene_names.\n\
+      By default will detect human and mouse mitochondrial genes from a gene symbol.\n"
+    info: null
+    default:
+    - "^[mM][tT]-"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_ribosomal_genes"
+    description: "In which .var slot to store a boolean array corresponding the ribosomal\
+      \ genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_ribosomal_fraction"
+    description: "When specified, write the fraction of counts originating from ribosomal\
+      \ genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified\
+      \ name.\nRequires --var_name_ribosomal_genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ribosomal_gene_regex"
+    description: "Regex string that identifies ribosomal genes from --var_gene_names.\n\
+      By default will detect human and mouse ribosomal genes from a gene symbol.\n"
+    info: null
+    default:
+    - "^[Mm]?[Rr][Pp][LlSs]"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "QC metrics calculation options"
+  arguments:
+  - type: "string"
+    name: "--var_qc_metrics"
+    description: "Keys to select a boolean (containing only True or False) column\
+      \ from .var.\nFor each cell, calculate the proportion of total values for genes\
+      \ which are labeled 'True', \ncompared to the total sum of the values for all\
+      \ genes. Defaults to the value from\n--var_name_mitochondrial_genes.\n"
+    info: null
+    example:
+    - "ercc,highly_variable"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+  - type: "integer"
+    name: "--top_n_vars"
+    description: "Number of top vars to be used to calculate cumulative proportions.\n\
+      If not specified, proportions are not calculated. `--top_n_vars 20,50` finds\n\
+      cumulative proportion to the 20th and 50th most expressed vars.\n"
+    info: null
+    default:
+    - 50
+    - 100
+    - 200
+    - 500
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+  - type: "string"
+    name: "--output_obs_num_nonzero_vars"
+    description: "Name of column in .obs describing, for each observation, the number\
+      \ of stored values\n(including explicit zeroes). In other words, the name of\
+      \ the column that counts\nfor each row the number of columns that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_vars"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_total_counts_vars"
+    description: "Name of the column for .obs describing, for each observation (row),\n\
+      the sum of the stored values in the columns.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_num_nonzero_obs"
+    description: "Name of column describing, for each feature, the number of stored\
+      \ values\n(including explicit zeroes). In other words, the name of the column\
+      \ that counts\nfor each column the number of rows that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_obs"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_total_counts_obs"
+    description: "Name of the column in .var describing, for each feature (column),\n\
+      the sum of the stored values in the rows.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_obs_mean"
+    description: "Name of the column in .obs providing the mean of the values in each\
+      \ row.\n"
+    info: null
+    default:
+    - "obs_mean"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_pct_dropout"
+    description: "Name of the column in .obs providing for each feature the percentage\
+      \ of\nobservations the feature does not appear on (i.e. is missing). Same as\
+      \ `--output_var_num_nonzero_obs`\nbut percentage based.\n"
+    info: null
+    default:
+    - "pct_dropout"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "A pipeline to add basic qc statistics to a MuData "
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "concat_test_data"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info:
+  test_dependencies:
+  - name: "qc_test"
+    namespace: "test_workflows/qc"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "metadata/grep_annotation_column"
+  repository:
+    type: "local"
+- name: "qc/calculate_qc_metrics"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/qc/qc/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/qc/qc"
+  executable: "target/nextflow/workflows/qc/qc/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "de02293c9e13198622b988dac952b2c8c70a1e35"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/metadata/grep_annotation_column"
+  - "target/nextflow/qc/calculate_qc_metrics"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.0"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.0'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/main.nf
new file mode 100644
index 0000000..53db6d5
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/main.nf
@@ -0,0 +1,3766 @@
+// qc v4.0.0
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author, maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "qc",
+  "namespace" : "workflows/qc",
+  "version" : "v4.0.0",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the sample.",
+          "example" : [
+            "foo"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to the sample.",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Layer to calculate qc metrics for.",
+          "example" : [
+            "raw_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Mitochondrial & Ribosomal Gene Detection",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_gene_names",
+          "description" : ".var column name to be used to detect mitochondrial/ribosomal genes instead of .var_names (default if not set).\nGene names matching with the regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will be \nidentified as mitochondrial or ribosomal genes, respectively.\n",
+          "example" : [
+            "gene_symbol"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_mitochondrial_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_mitochondrial_fraction",
+          "description" : ".Obs slot to store the fraction of reads found to be mitochondrial. Defaults to 'fraction_' suffixed by the value of --var_name_mitochondrial_genes\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--mitochondrial_gene_regex",
+          "description" : "Regex string that identifies mitochondrial genes from --var_gene_names.\nBy default will detect human and mouse mitochondrial genes from a gene symbol.\n",
+          "default" : [
+            "^[mM][tT]-"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_ribosomal_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_ribosomal_fraction",
+          "description" : "When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified name.\nRequires --var_name_ribosomal_genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--ribosomal_gene_regex",
+          "description" : "Regex string that identifies ribosomal genes from --var_gene_names.\nBy default will detect human and mouse ribosomal genes from a gene symbol.\n",
+          "default" : [
+            "^[Mm]?[Rr][Pp][LlSs]"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "QC metrics calculation options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_qc_metrics",
+          "description" : "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes. Defaults to the value from\n--var_name_mitochondrial_genes.\n",
+          "example" : [
+            "ercc,highly_variable"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        },
+        {
+          "type" : "integer",
+          "name" : "--top_n_vars",
+          "description" : "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated. `--top_n_vars 20,50` finds\ncumulative proportion to the 20th and 50th most expressed vars.\n",
+          "default" : [
+            50,
+            100,
+            200,
+            500
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_num_nonzero_vars",
+          "description" : "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each row the number of columns that contain data.\n",
+          "default" : [
+            "num_nonzero_vars"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_total_counts_vars",
+          "description" : "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_num_nonzero_obs",
+          "description" : "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each column the number of rows that contain data.\n",
+          "default" : [
+            "num_nonzero_obs"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_total_counts_obs",
+          "description" : "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_obs_mean",
+          "description" : "Name of the column in .obs providing the mean of the values in each row.\n",
+          "default" : [
+            "obs_mean"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_pct_dropout",
+          "description" : "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e. is missing). Same as `--output_var_num_nonzero_obs`\nbut percentage based.\n",
+          "default" : [
+            "pct_dropout"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "A pipeline to add basic qc statistics to a MuData ",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "info" : {
+    "test_dependencies" : [
+      {
+        "name" : "qc_test",
+        "namespace" : "test_workflows/qc"
+      }
+    ]
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "metadata/grep_annotation_column",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "qc/calculate_qc_metrics",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/qc/qc/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/qc/qc",
+    "viash_version" : "0.9.4",
+    "git_commit" : "de02293c9e13198622b988dac952b2c8c70a1e35",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.0",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.0'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { grep_annotation_column } from "${meta.resources_dir}/../../../../nextflow/metadata/grep_annotation_column/main.nf"
+include { calculate_qc_metrics } from "${meta.resources_dir}/../../../../nextflow/qc/calculate_qc_metrics/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+    output_ch = input_ch
+    // Avoid conflict between output from component and output for this workflow
+    | map {id, state -> 
+      assert state.output, "Output must be defined"
+      def new_state = state + ["workflow_output": state.output]
+      [id, new_state]
+    }
+    // Add default for var_qc_metrics component
+    | map {id, state ->
+      def var_qc_default = []
+      // Remove the var_qc_metric argument from the state if its value is null (not specified)
+      def new_state = state.findAll { it.key != "var_qc_metrics" || it.value == null }
+      if (state.var_name_mitochondrial_genes) {
+        var_qc_default.add(state.var_name_mitochondrial_genes)
+      }
+      if (state.var_name_ribosomal_genes) {
+        var_qc_default.add(state.var_name_ribosomal_genes)
+      }
+      // Get the new state, but make sure to overwrite var_qc_metrics if the user has set it.
+      new_state = ["var_qc_metrics": var_qc_default.join(";")] + new_state
+      [id, new_state]
+    }
+
+    | grep_annotation_column.run(
+      key: "grep_mitochondrial_genes",
+      runIf: { id, state ->
+        state.var_name_mitochondrial_genes
+      },
+      fromState: { id, state ->
+        def stateMapping = [
+          "input": state.input,
+          "modality": state.modality,
+          "input_column": state.var_gene_names,
+          "matrix": "var",
+          "output_match_column": state.var_name_mitochondrial_genes,
+          "regex_pattern": state.mitochondrial_gene_regex,
+          "input_layer": state.layer,
+        ]
+        stateMapping.output_fraction_column = state.obs_name_mitochondrial_fraction ? state.obs_name_mitochondrial_fraction: "fraction_$state.var_name_mitochondrial_genes"
+        return stateMapping
+      },
+      toState: ["input": "output"]
+    )
+
+    | grep_annotation_column.run(
+      key: "grep_ribosomal_genes",
+      runIf: { id, state ->
+        state.var_name_ribosomal_genes
+      },
+      fromState: { id, state ->
+        def stateMapping = [
+          "input": state.input,
+          "modality": state.modality,
+          "input_column": state.var_gene_names,
+          "matrix": "var",
+          "output_match_column": state.var_name_ribosomal_genes,
+          "regex_pattern": state.ribosomal_gene_regex,
+          "input_layer": state.layer,
+        ]
+        stateMapping.output_fraction_column = state.obs_name_ribosomal_fraction ? state.obs_name_ribosomal_fraction: "fraction_$state.var_name_ribosomal_genes"
+        return stateMapping
+      },
+      toState: ["input": "output"]
+    )
+
+    | calculate_qc_metrics.run(
+      fromState: { id, state ->
+        def newState = [
+          "input": state.input,
+          "modality": state.modality,
+          "layer": state.layer,
+          // TODO: remove this workaround when Viash issue is resolved:
+          //       'top_n_vars': list(map(int, r''.split(';'))),
+          //     ValueError: invalid literal for int() with base 10: ''
+          // See https://github.com/viash-io/viash/issues/619
+          "top_n_vars": state.top_n_vars ? state.top_n_vars : null,
+          "var_qc_metrics_fill_na_value": state.var_qc_metrics_fill_na_value,
+          "output_obs_num_nonzero_vars": state.output_obs_num_nonzero_vars,
+          "output_obs_total_counts_vars": state.output_obs_total_counts_vars,
+          "output_var_num_nonzero_obs": state.output_var_num_nonzero_obs,
+          "output_var_total_counts_obs": state.output_var_total_counts_obs,
+          "output_var_obs_mean": state.output_var_obs_mean,
+          "output_var_pct_dropout": state.output_var_pct_dropout,
+          "output": state.workflow_output,
+          "compression": "gzip"
+        ]
+        if (state.var_qc_metrics) {
+          newState += ["var_qc_metrics": state.var_qc_metrics]
+        }
+      return newState
+      },
+      toState: ["output": "output"]
+    )
+    | setState(["output"]) 
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/nextflow.config
new file mode 100644
index 0000000..d9abf28
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/qc/qc'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.0'
+  description = 'A pipeline to add basic qc statistics to a MuData '
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/nextflow_schema.json
new file mode 100644
index 0000000..82142b4
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/nextflow_schema.json
@@ -0,0 +1,190 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "qc",
+  "description": "A pipeline to add basic qc statistics to a MuData ",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "ID of the sample.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"foo\"`. "
+        },
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the sample.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality to process.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "layer": {
+          "type": "string",
+          "description": "Layer to calculate qc metrics for.",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"raw_counts\"`. "
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "mitochondrial & ribosomal gene detection": {
+      "title": "Mitochondrial & Ribosomal Gene Detection",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_gene_names": {
+          "type": "string",
+          "description": ".var column name to be used to detect mitochondrial/ribosomal genes instead of .var_names (default if not set).\nGene names matching with the regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will be \nidentified as mitochondrial or ribosomal genes, respectively.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gene_symbol\"`. "
+        },
+        "var_name_mitochondrial_genes": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_name_mitochondrial_fraction": {
+          "type": "string",
+          "description": ".Obs slot to store the fraction of reads found to be mitochondrial",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "mitochondrial_gene_regex": {
+          "type": "string",
+          "description": "Regex string that identifies mitochondrial genes from --var_gene_names.\nBy default will detect human and mouse mitochondrial genes from a gene symbol.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"^[mM][tT]-\"`. ",
+          "default": "^[mM][tT]-"
+        },
+        "var_name_ribosomal_genes": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_name_ribosomal_fraction": {
+          "type": "string",
+          "description": "When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified name.\nRequires --var_name_ribosomal_genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "ribosomal_gene_regex": {
+          "type": "string",
+          "description": "Regex string that identifies ribosomal genes from --var_gene_names.\nBy default will detect human and mouse ribosomal genes from a gene symbol.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"^[Mm]?[Rr][Pp][LlSs]\"`. ",
+          "default": "^[Mm]?[Rr][Pp][LlSs]"
+        }
+      }
+    },
+    "qc metrics calculation options": {
+      "title": "QC metrics calculation options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_qc_metrics": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"ercc,highly_variable\"]`. "
+        },
+        "top_n_vars": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated",
+          "help_text": "Type: `integer`, multiple: `True`, default: `[50,100,200,500]`. ",
+          "default": [
+            50,
+            100,
+            200,
+            500
+          ]
+        },
+        "output_obs_num_nonzero_vars": {
+          "type": "string",
+          "description": "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes)",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"num_nonzero_vars\"`. ",
+          "default": "num_nonzero_vars"
+        },
+        "output_obs_total_counts_vars": {
+          "type": "string",
+          "description": "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"total_counts\"`. ",
+          "default": "total_counts"
+        },
+        "output_var_num_nonzero_obs": {
+          "type": "string",
+          "description": "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes)",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"num_nonzero_obs\"`. ",
+          "default": "num_nonzero_obs"
+        },
+        "output_var_total_counts_obs": {
+          "type": "string",
+          "description": "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"total_counts\"`. ",
+          "default": "total_counts"
+        },
+        "output_var_obs_mean": {
+          "type": "string",
+          "description": "Name of the column in .obs providing the mean of the values in each row.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"obs_mean\"`. ",
+          "default": "obs_mean"
+        },
+        "output_var_pct_dropout": {
+          "type": "string",
+          "description": "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"pct_dropout\"`. ",
+          "default": "pct_dropout"
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/mitochondrial & ribosomal gene detection"
+    },
+    {
+      "$ref": "#/$defs/qc metrics calculation options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/.config.vsh.yaml
new file mode 100644
index 0000000..a8d3841
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/.config.vsh.yaml
@@ -0,0 +1,336 @@
+name: "move_mudata_obs_to_tiledb"
+namespace: "tiledb"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Input database"
+  description: "Open a tileDB-SOMA database by URI or as a local directory."
+  arguments:
+  - type: "string"
+    name: "--input_uri"
+    description: "A URI pointing to a TileDB-SOMA database. Mutually exclusive with\
+      \ 'input_dir'"
+    info: null
+    example:
+    - "s3://bucket/path"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input_dir"
+    description: "Path to a TileDB-SOMA database as a local directory"
+    info: null
+    example:
+    - "./tiledb_database"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--s3_region"
+    description: "Region where the TileDB-SOMA database is hosted.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--endpoint"
+    description: "Custom endpoint to use to connect to S3\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--s3_no_sign_request"
+    description: "Do not sign S3 requests. Credentials will not be loaded if this\
+      \ argument is provided.\n"
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_modality"
+    description: "TileDB-SOMA measurement to add the output to.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_index_name_input"
+    description: "Name of the index that is used to describe the cells (observations).\n"
+    info: null
+    default:
+    - "cell_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "MuData input"
+  arguments:
+  - type: "file"
+    name: "--input_mudata"
+    description: "MuData object to take the columns from. The observations and their\
+      \ order should\nmatch between the database and the input modality.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Modality where to take the .obs from.\n"
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_input"
+    description: "Columns from .obs to copy. The keys should not be present yet in\
+      \ the database.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "TileDB-SOMA output"
+  arguments:
+  - type: "file"
+    name: "--output_tiledb"
+    description: "Output to a directory instead of adding to the existing database.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Move .obs columns from a MuData modality to an existing tileDB database.\n\
+  The .obs keys should not exist in the database yet; and the observations from the\
+  \ modality and \ntheir order should match with what is already present the tiledb\
+  \ database.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "tiledb"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "private"
+  target: "private"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+  docker_run_args:
+  - "--env"
+  - "AWS_ACCESS_KEY_ID"
+  - "--env"
+  - "AWS_SECRET_ACCESS_KEY"
+  - "--env"
+  - "AWS_DEFAULT_REGION"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "highmem"
+    - "midcpu"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "tiledbsoma"
+    - "boto3"
+    - "awscli"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  - type: "python"
+    user: false
+    packages:
+    - "moto[server]"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/tiledb/move_mudata_obs_to_tiledb/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/_private/nextflow/tiledb/move_mudata_obs_to_tiledb"
+  executable: "target/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/main.nf
new file mode 100644
index 0000000..5396be4
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/main.nf
@@ -0,0 +1,4138 @@
+// move_mudata_obs_to_tiledb v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author, maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "move_mudata_obs_to_tiledb",
+  "namespace" : "tiledb",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Input database",
+      "description" : "Open a tileDB-SOMA database by URI or as a local directory.",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--input_uri",
+          "description" : "A URI pointing to a TileDB-SOMA database. Mutually exclusive with 'input_dir'",
+          "example" : [
+            "s3://bucket/path"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input_dir",
+          "description" : "Path to a TileDB-SOMA database as a local directory",
+          "example" : [
+            "./tiledb_database"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--s3_region",
+          "description" : "Region where the TileDB-SOMA database is hosted.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--endpoint",
+          "description" : "Custom endpoint to use to connect to S3\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--s3_no_sign_request",
+          "description" : "Do not sign S3 requests. Credentials will not be loaded if this argument is provided.\n",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_modality",
+          "description" : "TileDB-SOMA measurement to add the output to.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_index_name_input",
+          "description" : "Name of the index that is used to describe the cells (observations).\n",
+          "default" : [
+            "cell_id"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "MuData input",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input_mudata",
+          "description" : "MuData object to take the columns from. The observations and their order should\nmatch between the database and the input modality.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Modality where to take the .obs from.\n",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_input",
+          "description" : "Columns from .obs to copy. The keys should not be present yet in the database.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "TileDB-SOMA output",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output_tiledb",
+          "description" : "Output to a directory instead of adding to the existing database.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Move .obs columns from a MuData modality to an existing tileDB database.\nThe .obs keys should not exist in the database yet; and the observations from the modality and \ntheir order should match with what is already present the tiledb database.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/tiledb/"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "private",
+    "target" : "private"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild",
+      "docker_run_args" : [
+        "--env",
+        "AWS_ACCESS_KEY_ID",
+        "--env",
+        "AWS_SECRET_ACCESS_KEY",
+        "--env",
+        "AWS_DEFAULT_REGION"
+      ]
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "highmem",
+          "midcpu"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "tiledbsoma",
+            "boto3",
+            "awscli"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "moto[server]"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/tiledb/move_mudata_obs_to_tiledb/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/_private/nextflow/tiledb/move_mudata_obs_to_tiledb",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+import mudata
+import tiledbsoma
+import tiledbsoma.io
+import pandas as pd
+from pathlib import Path
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input_uri': $( if [ ! -z ${VIASH_PAR_INPUT_URI+x} ]; then echo "r'${VIASH_PAR_INPUT_URI//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_dir': $( if [ ! -z ${VIASH_PAR_INPUT_DIR+x} ]; then echo "r'${VIASH_PAR_INPUT_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  's3_region': $( if [ ! -z ${VIASH_PAR_S3_REGION+x} ]; then echo "r'${VIASH_PAR_S3_REGION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'endpoint': $( if [ ! -z ${VIASH_PAR_ENDPOINT+x} ]; then echo "r'${VIASH_PAR_ENDPOINT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  's3_no_sign_request': $( if [ ! -z ${VIASH_PAR_S3_NO_SIGN_REQUEST+x} ]; then echo "r'${VIASH_PAR_S3_NO_SIGN_REQUEST//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'output_modality': $( if [ ! -z ${VIASH_PAR_OUTPUT_MODALITY+x} ]; then echo "r'${VIASH_PAR_OUTPUT_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_index_name_input': $( if [ ! -z ${VIASH_PAR_OBS_INDEX_NAME_INPUT+x} ]; then echo "r'${VIASH_PAR_OBS_INDEX_NAME_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_mudata': $( if [ ! -z ${VIASH_PAR_INPUT_MUDATA+x} ]; then echo "r'${VIASH_PAR_INPUT_MUDATA//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_input': $( if [ ! -z ${VIASH_PAR_OBS_INPUT+x} ]; then echo "r'${VIASH_PAR_OBS_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'output_tiledb': $( if [ ! -z ${VIASH_PAR_OUTPUT_TILEDB+x} ]; then echo "r'${VIASH_PAR_OUTPUT_TILEDB//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+tiledbsoma.logging.info()
+
+
+def download_s3_dir(input_uri, output_dir_path):
+    logger.info("Requested to output to a directory. Downloading database...")
+    output_dir_path.mkdir(parents=True, exist_ok=True)
+    import boto3
+    from awscli.customizations.s3.utils import split_s3_bucket_key
+
+    bucket, key = split_s3_bucket_key(input_uri)
+    connection_args = {
+        "endpoint_url": par["endpoint"],
+        "region_name": par["s3_region"],
+    }
+    if par["s3_no_sign_request"]:
+        import botocore
+
+        connection_args["config"] = botocore.config.Config(
+            signature_version=botocore.UNSIGNED
+        )
+
+    client = boto3.resource("s3", **connection_args)
+    bucket = client.Bucket(bucket)
+    for i, s3_obj in enumerate(bucket.objects.filter(Prefix=key)):
+        output_path = output_dir_path / s3_obj.key.removeprefix(key).lstrip("/")
+        output_path.parent.mkdir(parents=True, exist_ok=True)
+        bucket.download_file(s3_obj.key, output_path)
+        print(f"Downloaded {i} files.", file=sys.stdout, flush=True, end="\\\\r")
+    logger.info("Download completed!")
+
+
+def main(par):
+    logger.info(f"Component {meta['name']} started.")
+    if par["input_uri"]:
+        par["input_uri"] = par["input_uri"].rstrip("/")
+    if par["input_uri"] and par["input_dir"]:
+        raise ValueError("Cannot provide both 'input_uri' and 'input_dir'.")
+    if not par["input_uri"] and not par["input_dir"]:
+        raise ValueError("Must provide either 'input_uri' or 'input_dir'")
+    if not par["obs_input"]:
+        raise ValueError("Please provide at least one .obs column.")
+    logger.info(
+        "Opening mudata file '%s', modality '%s'.", par["input_mudata"], par["modality"]
+    )
+    modality_data = mudata.read_h5ad(par["input_mudata"], mod=par["modality"])
+    logger.info(
+        "Done reading modality. Looking at .obs for keys: '%s'",
+        ",".join(par["obs_input"]),
+    )
+    try:
+        for obs_key in par["obs_input"]:
+            modality_data.obs[obs_key]
+    except KeyError as e:
+        raise KeyError("Not all .obs keys were found in the input!") from e
+
+    logger.info("Done getting .obs keys.")
+    optional_config = {
+        "vfs.s3.region": par["s3_region"],
+        "vfs.s3.endpoint_override": par["endpoint"],
+        "vfs.s3.no_sign_request": par["s3_no_sign_request"],
+    }
+    tiledb_config = {}
+    for config_setting, config_val in optional_config.items():
+        if config_val is not None:
+            tiledb_config[config_setting] = config_val
+    logger.info("Using the following config to connect to S3: %s", tiledb_config)
+
+    if par["output_tiledb"]:
+        output_dir_path = Path(par["output_tiledb"])
+        if par["input_dir"]:
+            import shutil
+
+            shutil.copytree(par["input_dir"], output_dir_path, dirs_exist_ok=True)
+        else:
+            download_s3_dir(par["input_uri"], output_dir_path)
+        logger.info("Setting input to '%s'", output_dir_path)
+        par["input_uri"] = f"file://{output_dir_path.resolve()}"
+        logger.info("Overwriting TileDB config because S3 connection is not required.")
+        tiledb_config = {}
+
+    logger.info("Trying to access '%s'", par["input_uri"])
+    logger.info("Fetching .obs")
+    with tiledbsoma.open(
+        par["input_uri"],
+        mode="r",
+        context=tiledbsoma.SOMATileDBContext(tiledb_config=tiledb_config),
+    ) as open_experiment:
+        logger.info("Connection established.")
+        obs_df = open_experiment.obs.read().concat().to_pandas()
+        logger.info("Done downloading .obs from databse.")
+    logger.info("Adding obs columns to fetched .obs dataframe.")
+    overlapping_obs = set(par["obs_input"]).intersection(set(obs_df.columns.to_list()))
+    if overlapping_obs:
+        raise ValueError(
+            f"The following keys already exist in the database: {','.join(overlapping_obs)}."
+        )
+
+    columns_to_add = modality_data.obs[par["obs_input"]]
+    new_obs = pd.merge(
+        obs_df,
+        columns_to_add,
+        left_on=par["obs_index_name_input"],
+        right_index=True,
+        how="right",
+    )
+    logger.info(
+        "Writing obs back to database. Connection to %s with config %s",
+        par["input_uri"],
+        tiledb_config,
+    )
+    with tiledbsoma.open(
+        par["input_uri"],
+        mode="w",
+        context=tiledbsoma.SOMATileDBContext(tiledb_config=tiledb_config),
+    ) as open_experiment:
+        tiledbsoma.io.update_obs(
+            open_experiment,
+            new_data=new_obs,
+            default_index_name=par["obs_index_name_input"],
+            context=tiledbsoma.SOMATileDBContext(tiledb_config=tiledb_config),
+        )
+    logger.info("Finished!")
+
+
+if __name__ == "__main__":
+    main(par)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/tiledb/move_mudata_obs_to_tiledb",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "highmem",
+    "midcpu"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/nextflow.config
new file mode 100644
index 0000000..06ab553
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'tiledb/move_mudata_obs_to_tiledb'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Move .obs columns from a MuData modality to an existing tileDB database.\nThe .obs keys should not exist in the database yet; and the observations from the modality and \ntheir order should match with what is already present the tiledb database.\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obs_to_tiledb/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/.config.vsh.yaml
new file mode 100644
index 0000000..75eb6ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/.config.vsh.yaml
@@ -0,0 +1,313 @@
+name: "move_mudata_obsm_to_tiledb"
+namespace: "tiledb"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Input database"
+  description: "Open a tileDB-SOMA database by URI."
+  arguments:
+  - type: "string"
+    name: "--input_uri"
+    description: "A URI pointing to a TileDB-SOMA database."
+    info: null
+    example:
+    - "s3://bucket/path"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--s3_region"
+    description: "Region where the TileDB-SOMA database is hosted.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--endpoint"
+    description: "Custom endpoint to use to connect to S3\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--s3_no_sign_request"
+    description: "Do not sign S3 requests. Credentials will not be loaded if this\
+      \ argument is provided.\n"
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_modality"
+    description: "TileDB-SOMA measurement to add the output to.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "MuData input"
+  arguments:
+  - type: "file"
+    name: "--input_mudata"
+    description: "MuData object to take the columns from. The observations and their\
+      \ order should\nmatch between the database and the input modality.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Modality where to take the .obsm from.\n"
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_input"
+    description: "Keys from .obm to copy. The keys should not be present yet in the\
+      \ database.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "TileDB-SOMA output"
+  arguments:
+  - type: "file"
+    name: "--output_tiledb"
+    description: "Output to a directory instead of adding to the existing database.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Move .obsm items from a MuData modality to an existing tileDB database.\n\
+  The .obsm keys should not exist in the database yet; and the observations from the\
+  \ modality and \ntheir order should match with what is already present the tiledb\
+  \ database.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "tiledb"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "private"
+  target: "private"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+  docker_run_args:
+  - "--env"
+  - "AWS_ACCESS_KEY_ID"
+  - "--env"
+  - "AWS_SECRET_ACCESS_KEY"
+  - "--env"
+  - "AWS_DEFAULT_REGION"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "highmem"
+    - "midcpu"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "tiledbsoma"
+    - "boto3"
+    - "awscli"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  - type: "python"
+    user: false
+    packages:
+    - "moto[server]"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/tiledb/move_mudata_obsm_to_tiledb/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb"
+  executable: "target/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/main.nf
new file mode 100644
index 0000000..42a95e0
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/main.nf
@@ -0,0 +1,4103 @@
+// move_mudata_obsm_to_tiledb v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author, maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "move_mudata_obsm_to_tiledb",
+  "namespace" : "tiledb",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Input database",
+      "description" : "Open a tileDB-SOMA database by URI.",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--input_uri",
+          "description" : "A URI pointing to a TileDB-SOMA database.",
+          "example" : [
+            "s3://bucket/path"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--s3_region",
+          "description" : "Region where the TileDB-SOMA database is hosted.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--endpoint",
+          "description" : "Custom endpoint to use to connect to S3\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--s3_no_sign_request",
+          "description" : "Do not sign S3 requests. Credentials will not be loaded if this argument is provided.\n",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_modality",
+          "description" : "TileDB-SOMA measurement to add the output to.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "MuData input",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input_mudata",
+          "description" : "MuData object to take the columns from. The observations and their order should\nmatch between the database and the input modality.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Modality where to take the .obsm from.\n",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_input",
+          "description" : "Keys from .obm to copy. The keys should not be present yet in the database.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "TileDB-SOMA output",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output_tiledb",
+          "description" : "Output to a directory instead of adding to the existing database.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Move .obsm items from a MuData modality to an existing tileDB database.\nThe .obsm keys should not exist in the database yet; and the observations from the modality and \ntheir order should match with what is already present the tiledb database.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/tiledb/"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "private",
+    "target" : "private"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild",
+      "docker_run_args" : [
+        "--env",
+        "AWS_ACCESS_KEY_ID",
+        "--env",
+        "AWS_SECRET_ACCESS_KEY",
+        "--env",
+        "AWS_DEFAULT_REGION"
+      ]
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "highmem",
+          "midcpu"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "tiledbsoma",
+            "boto3",
+            "awscli"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "moto[server]"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/tiledb/move_mudata_obsm_to_tiledb/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+import mudata
+import tiledbsoma
+import tiledbsoma.io
+import pandas as pd
+import json
+from pathlib import Path
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input_uri': $( if [ ! -z ${VIASH_PAR_INPUT_URI+x} ]; then echo "r'${VIASH_PAR_INPUT_URI//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  's3_region': $( if [ ! -z ${VIASH_PAR_S3_REGION+x} ]; then echo "r'${VIASH_PAR_S3_REGION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'endpoint': $( if [ ! -z ${VIASH_PAR_ENDPOINT+x} ]; then echo "r'${VIASH_PAR_ENDPOINT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  's3_no_sign_request': $( if [ ! -z ${VIASH_PAR_S3_NO_SIGN_REQUEST+x} ]; then echo "r'${VIASH_PAR_S3_NO_SIGN_REQUEST//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'output_modality': $( if [ ! -z ${VIASH_PAR_OUTPUT_MODALITY+x} ]; then echo "r'${VIASH_PAR_OUTPUT_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_mudata': $( if [ ! -z ${VIASH_PAR_INPUT_MUDATA+x} ]; then echo "r'${VIASH_PAR_INPUT_MUDATA//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_input': $( if [ ! -z ${VIASH_PAR_OBSM_INPUT+x} ]; then echo "r'${VIASH_PAR_OBSM_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'output_tiledb': $( if [ ! -z ${VIASH_PAR_OUTPUT_TILEDB+x} ]; then echo "r'${VIASH_PAR_OUTPUT_TILEDB//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+tiledbsoma.logging.info()
+
+
+def main(par):
+    logger.info(f"Component {meta['name']} started.")
+    par["input_uri"] = par["input_uri"].rstrip("/")
+    if not par["obsm_input"]:
+        raise ValueError("Please provide at least one .obsm key.")
+    logger.info(
+        "Opening mudata file '%s', modality '%s'.", par["input_mudata"], par["modality"]
+    )
+    modality_data = mudata.read_h5ad(par["input_mudata"], mod=par["modality"])
+    logger.info(
+        "Done reading modality. Looking at .obsm for keys: '%s'",
+        ",".join(par["obsm_input"]),
+    )
+    try:
+        keys_to_transfer = {
+            obsm_key: modality_data.obsm[obsm_key] for obsm_key in par["obsm_input"]
+        }
+    except KeyError as e:
+        raise KeyError("Not all .obsm keys were found in the input!") from e
+
+    logger.info("Done getting .obsm keys.")
+    optional_config = {
+        "vfs.s3.region": par["s3_region"],
+        "vfs.s3.endpoint_override": par["endpoint"],
+        "vfs.s3.no_sign_request": par["s3_no_sign_request"],
+    }
+    tiledb_config = {}
+    for config_setting, config_val in optional_config.items():
+        if config_val is not None:
+            tiledb_config[config_setting] = config_val
+    logger.info("Using the following config to connect to S3: %s", tiledb_config)
+    context = tiledbsoma.SOMATileDBContext(tiledb_config=tiledb_config)
+
+    if par["output_tiledb"]:
+        logger.info("Requested to output to a directory. Downloading database...")
+        output_dir_path = Path(par["output_tiledb"])
+        output_dir_path.mkdir(parents=True, exist_ok=True)
+        import boto3
+        from awscli.customizations.s3.utils import split_s3_bucket_key
+
+        bucket, key = split_s3_bucket_key(par["input_uri"])
+        connection_args = {
+            "endpoint_url": par["endpoint"],
+            "region_name": par["s3_region"],
+        }
+        if par["s3_no_sign_request"]:
+            import botocore
+
+            connection_args["config"] = botocore.config.Config(
+                signature_version=botocore.UNSIGNED
+            )
+
+        client = boto3.resource("s3", **connection_args)
+        bucket = client.Bucket(bucket)
+        for i, s3_obj in enumerate(bucket.objects.filter(Prefix=key)):
+            output_path = output_dir_path / s3_obj.key.removeprefix(key).lstrip("/")
+            output_path.parent.mkdir(parents=True, exist_ok=True)
+            bucket.download_file(s3_obj.key, output_path)
+            print(f"Downloaded {i} files.", file=sys.stdout, flush=True, end="\\\\r")
+        logger.info("Download completed!")
+        logger.info("Setting input to %s", par["input_uri"])
+        par["input_uri"] = f"file://{output_dir_path.resolve()}"
+        logger.info("Overwriting TileDB config because S3 connection is not required.")
+        context = tiledbsoma.SOMATileDBContext()
+
+    logger.info("Trying to access '%s'", par["input_uri"])
+    with tiledbsoma.open(
+        par["input_uri"], mode="w", context=context
+    ) as open_experiment:
+        logger.info("Connection established.")
+        logger.info("Looking for measurement %s", par["output_modality"])
+        measurement = open_experiment.ms[par["output_modality"]]
+        logger.info("Checking if keys do not already exist.")
+        existing_keys = measurement.obsm.keys()
+        logger.info("Existing keys: %s", ",".join(existing_keys))
+        overlap = set(existing_keys).intersection(set(keys_to_transfer.keys()))
+        if overlap:
+            raise ValueError(
+                f"The following keys already exist in the database: {','.join(overlap)}."
+            )
+        logger.info("Adding keys to database.")
+        for key, obsm_val in keys_to_transfer.items():
+            logger.info("Adding .obsm key '%s', of class '%s'", key, type(obsm_val))
+            index_as_json = None
+            if isinstance(obsm_val, pd.DataFrame):
+                # tileDB does not allow column indices to be saved directly
+                # So need to add those as JSON metadata
+                index_to_write = obsm_val.columns.to_list()
+                if not isinstance(index_to_write, pd.RangeIndex):
+                    index_as_json = json.dumps(index_to_write)
+                obsm_val = obsm_val.to_numpy()
+
+            tiledbsoma.io.add_matrix_to_collection(
+                open_experiment,
+                measurement_name=par["output_modality"],
+                ingest_mode="write",
+                collection_name="obsm",
+                matrix_name=key,
+                matrix_data=obsm_val,
+                context=context,
+            )
+            # Allow for more than one obsm slot to be transferred
+            open_experiment = open_experiment.reopen(mode="w")
+            if index_as_json:
+                uri = f"{par['input_uri']}/ms/{par['output_modality']}/obsm/{key}"
+                with tiledbsoma.open(
+                    uri=uri, mode="w", context=context
+                ) as open_obsm_array:
+                    open_obsm_array.metadata["column_index"] = index_as_json
+
+    logger.info("Finished!")
+
+
+if __name__ == "__main__":
+    main(par)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/tiledb/move_mudata_obsm_to_tiledb",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "highmem",
+    "midcpu"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/nextflow.config
new file mode 100644
index 0000000..8213686
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'tiledb/move_mudata_obsm_to_tiledb'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Move .obsm items from a MuData modality to an existing tileDB database.\nThe .obsm keys should not exist in the database yet; and the observations from the modality and \ntheir order should match with what is already present the tiledb database.\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/.config.vsh.yaml
new file mode 100644
index 0000000..60560fb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/.config.vsh.yaml
@@ -0,0 +1,233 @@
+name: "split_modalities"
+namespace: "workflows/multiomics"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the sample."
+    info: null
+    example:
+    - "foo"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to the sample."
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output directory containing multiple h5mu files."
+    info: null
+    example:
+    - "/path/to/output"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_types"
+    description: "A csv containing the base filename and modality type per output\
+      \ file."
+    info: null
+    example:
+    - "types.csv"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "A pipeline to split a multimodal mudata files into several unimodal\
+  \ mudata files."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu"
+info:
+  test_dependencies:
+  - name: "split_modalities_test"
+    namespace: "test_workflows/multiomics"
+status: "enabled"
+scope:
+  image: "private"
+  target: "private"
+dependencies:
+- name: "dataflow/split_modalities"
+  alias: "split_modalities_component"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/multiomics/split_modalities/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/_private/nextflow/workflows/multiomics/split_modalities"
+  executable: "target/_private/nextflow/workflows/multiomics/split_modalities/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/dataflow/split_modalities"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/main.nf
new file mode 100644
index 0000000..4e59dc7
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/main.nf
@@ -0,0 +1,3521 @@
+// split_modalities v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author, maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "split_modalities",
+  "namespace" : "workflows/multiomics",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the sample.",
+          "example" : [
+            "foo"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to the sample.",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output directory containing multiple h5mu files.",
+          "example" : [
+            "/path/to/output"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_types",
+          "description" : "A csv containing the base filename and modality type per output file.",
+          "example" : [
+            "types.csv"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "A pipeline to split a multimodal mudata files into several unimodal mudata files.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu"
+    }
+  ],
+  "info" : {
+    "test_dependencies" : [
+      {
+        "name" : "split_modalities_test",
+        "namespace" : "test_workflows/multiomics"
+      }
+    ]
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "private",
+    "target" : "private"
+  },
+  "dependencies" : [
+    {
+      "name" : "dataflow/split_modalities",
+      "alias" : "split_modalities_component",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/multiomics/split_modalities/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/_private/nextflow/workflows/multiomics/split_modalities",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { split_modalities as split_modalities_component_viashalias } from "${meta.resources_dir}/../../../../../nextflow/dataflow/split_modalities/main.nf"
+split_modalities_component = split_modalities_component_viashalias.run(key: "split_modalities_component")
+
+// inner workflow
+// user-provided Nextflow code
+process splitStub {
+  input:
+    tuple val(id), val(unused)
+
+  output:
+    tuple val(id), path("stub_h5mus"), path("modalities.csv")
+
+  script:
+    """
+    echo "This process is not meant to be run without -stub being defined."
+    exit 1
+    """
+
+  stub:
+    """
+    mkdir stub_h5mus
+    touch stub_h5mus/${id}_vdj.h5mu
+    touch stub_h5mus/${id}_rna.h5mu
+    touch stub_h5mus/${id}_prot.h5mu
+    echo -e "name,filename\nrna,stub_h5mus/${id}_rna.h5mu\nprot,stub_h5mus/${id}_prot.h5mu\nvdj,stub_h5mus/${id}_vdj.h5mu" > modalities.csv
+    """
+}
+
+workflow run_wf {
+  // Split multimodal MuData files into several unimodal MuData files.
+  take:
+    input_ch
+
+  main:
+    split_ch = input_ch
+      | split_modalities_component.run(
+        filter: {!workflow.stubRun}, 
+        fromState: ["input": "input"],
+        toState: [
+          "output": "output",
+          "output_types": "output_types"
+        ]
+      )
+
+    split_stub_ch = input_ch
+      | filter{workflow.stubRun}
+      // This is not a build viash component, so we cannot use
+      // fromState or toState functionality
+      | map {id, state -> [id, state.input]}
+      | splitStub
+      | map {id, output, output_types ->
+        [id, ["output": output, "output_types": output_types]]
+      }
+
+    output_ch = split_ch.concat(split_stub_ch)
+      | setState(["output", "output_types"])
+
+  emit:
+    output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/nextflow.config
new file mode 100644
index 0000000..6a324bc
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/multiomics/split_modalities'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'A pipeline to split a multimodal mudata files into several unimodal mudata files.'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/multiomics/split_modalities/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/.config.vsh.yaml
new file mode 100644
index 0000000..7531aff
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/.config.vsh.yaml
@@ -0,0 +1,250 @@
+name: "log_normalize"
+namespace: "workflows/rna"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "MuData file to transform."
+    info: null
+    example:
+    - "dataset.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Modality to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Input layer containing raw counts. If not specified, .X is used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Transormation options"
+  arguments:
+  - type: "integer"
+    name: "--target_sum"
+    description: "Normalize total counts to the specified amount. If not set, after\
+      \ normalization each observation (cell) \nwill have a total count equal to the\
+      \ median of total counts for observations (cells) before normalization.\n"
+    info: null
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Output slots"
+  arguments:
+  - type: "string"
+    name: "--output_layer"
+    description: "Layer to write the log-transformed counts to.\n"
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Output"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Performs normalization and subsequent log-transformation of raw count\
+  \ data."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "private"
+  target: "private"
+dependencies:
+- name: "transform/normalize_total"
+  repository:
+    type: "local"
+- name: "transform/log1p"
+  repository:
+    type: "local"
+- name: "transform/delete_layer"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/rna/log_normalize/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/_private/nextflow/workflows/rna/log_normalize"
+  executable: "target/_private/nextflow/workflows/rna/log_normalize/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/transform/normalize_total"
+  - "target/nextflow/transform/log1p"
+  - "target/nextflow/transform/delete_layer"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/main.nf
new file mode 100644
index 0000000..33ebaff
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/main.nf
@@ -0,0 +1,3542 @@
+// log_normalize v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "log_normalize",
+  "namespace" : "workflows/rna",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "MuData file to transform.",
+          "example" : [
+            "dataset.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Modality to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Input layer containing raw counts. If not specified, .X is used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Transormation options",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--target_sum",
+          "description" : "Normalize total counts to the specified amount. If not set, after normalization each observation (cell) \nwill have a total count equal to the median of total counts for observations (cells) before normalization.\n",
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Output slots",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--output_layer",
+          "description" : "Layer to write the log-transformed counts to.\n",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Output",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Performs normalization and subsequent log-transformation of raw count data.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "private",
+    "target" : "private"
+  },
+  "dependencies" : [
+    {
+      "name" : "transform/normalize_total",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "transform/log1p",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "transform/delete_layer",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/rna/log_normalize/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/_private/nextflow/workflows/rna/log_normalize",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { normalize_total } from "${meta.resources_dir}/../../../../../nextflow/transform/normalize_total/main.nf"
+include { log1p } from "${meta.resources_dir}/../../../../../nextflow/transform/log1p/main.nf"
+include { delete_layer } from "${meta.resources_dir}/../../../../../nextflow/transform/delete_layer/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+
+
+
+workflow run_wf {
+    take: 
+    input_ch
+        
+    main:
+    output_ch = input_ch
+      | normalize_total.run(
+        fromState: [
+          "input": "input",
+          "modality": "modality",
+          "input_layer": "layer",
+          "target_sum": "target_sum"
+        ],
+        args: [
+          "output_layer": "normalized",
+        ],
+        toState: [
+          "input": "output",
+        ]
+      )
+      | log1p.run( 
+        fromState: [
+          "input": "input",
+          "modality": "modality",
+          "output_layer": "output_layer"
+        ],
+        args: [
+          "input_layer": "normalized",
+        ],
+        toState: [
+          "input": "output"
+        ]
+      )
+      | delete_layer.run(
+        fromState: [
+          "input": "input",
+          "modality": "modality"
+        ],
+        args: [
+          "layer": "normalized"
+        ],
+        toState: [
+          "output": "output"
+        ]
+      )
+      | setState(["output"])
+
+    emit:
+    output_ch
+
+
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/nextflow.config
new file mode 100644
index 0000000..94415d2
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/rna/log_normalize'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Performs normalization and subsequent log-transformation of raw count data.'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/_private/nextflow/workflows/rna/log_normalize/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/.config.vsh.yaml
new file mode 100644
index 0000000..9433c9e
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/.config.vsh.yaml
@@ -0,0 +1,479 @@
+name: "celltypist"
+namespace: "annotate"
+version: "v4.0.4"
+authors:
+- name: "Jakub Majercik"
+  roles:
+  - "author"
+  info:
+    role: "Contributor"
+    links:
+      email: "jakub@data-intuitive.com"
+      github: "jakubmajercik"
+      linkedin: "jakubmajercik"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Bioinformatics Engineer"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Inputs"
+  description: "Input dataset (query) arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "The input (query) data to be labeled. Should be a .h5mu file."
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_layer"
+    description: "The layer in the input data containing counts that are lognormalized\
+      \ to 10000, .X is not to be used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_var_gene_names"
+    description: "The name of the adata var column in the input data containing gene\
+      \ names; when no gene_name_layer is provided, the var index will be used.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--input_reference_gene_overlap"
+    description: "The minimum number of genes present in both the reference and query\
+      \ datasets.\n"
+    info: null
+    default:
+    - 100
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--sanitize_ensembl_ids"
+    description: "Whether to sanitize ensembl ids by removing version numbers."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Reference"
+  description: "Arguments related to the reference dataset."
+  arguments:
+  - type: "file"
+    name: "--reference"
+    description: "The reference data to train the CellTypist classifiers on. Only\
+      \ required if a pre-trained --model is not provided."
+    info: null
+    example:
+    - "reference.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_layer"
+    description: "The layer in the reference data containing counts that are lognormalized\
+      \ to 10000, if .X is not to be used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_obs_target"
+    description: "The name of the adata obs column in the reference data containing\
+      \ cell type annotations."
+    info: null
+    default:
+    - "cell_ontology_class"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_var_input"
+    description: ".var column containing highly variable genes. By default, do not\
+      \ subset genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_var_gene_names"
+    description: "The name of the adata var column in the reference data containing\
+      \ gene names; when no gene_name_layer is provided, the var index will be used.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Model arguments"
+  description: "Model arguments."
+  arguments:
+  - type: "file"
+    name: "--model"
+    description: "Pretrained model in pkl format. If not provided, the model will\
+      \ be trained on the reference data and --reference should be provided."
+    info: null
+    example:
+    - "pretrained_model.pkl"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--feature_selection"
+    description: "Whether to perform feature selection."
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--majority_voting"
+    description: "Whether to refine the predicted labels by running the majority voting\
+      \ classifier after over-clustering."
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--C"
+    description: "Inverse of regularization strength in logistic regression."
+    info: null
+    default:
+    - 1.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_iter"
+    description: "Maximum number of iterations before reaching the minimum of the\
+      \ cost function."
+    info: null
+    default:
+    - 1000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--use_SGD"
+    description: "Whether to use the stochastic gradient descent algorithm."
+    info: null
+    direction: "input"
+  - type: "double"
+    name: "--min_prop"
+    description: "\"For the dominant cell type within a subcluster, the minimum proportion\
+      \ of cells required to \nsupport naming of the subcluster by this cell type.\
+      \ Ignored if majority_voting is set to False. \nSubcluster that fails to pass\
+      \ this proportion threshold will be assigned 'Heterogeneous'.\"\n"
+    info: null
+    default:
+    - 0.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  description: "Output arguments."
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Output h5mu file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_predictions"
+    description: "In which `.obs` slots to store the predicted information.\n"
+    info: null
+    default:
+    - "celltypist_pred"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_probability"
+    description: "In which `.obs` slots to store the probability of the predictions.\n"
+    info: null
+    default:
+    - "celltypist_probability"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "cross_check_genes.py"
+- type: "file"
+  path: "subset_vars.py"
+- type: "file"
+  path: "set_var_index.py"
+- type: "file"
+  path: "is_lognormalized.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Automated cell type annotation tool for scRNA-seq datasets on the basis\
+  \ of logistic regression classifiers optimised by the stochastic gradient descent\
+  \ algorithm."
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "annotation_test_data"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "highcpu"
+    - "highmem"
+    - "highdisk"
+    - "gpu"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "nvcr.io/nvidia/pytorch:25.11-py3"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "python"
+    user: false
+    packages:
+    - "scanpy~=1.11.4"
+    upgrade: true
+  - type: "python"
+    user: false
+    packages:
+    - "celltypist==1.7.1"
+    upgrade: true
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/annotate/celltypist/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/annotate/celltypist"
+  executable: "target/nextflow/annotate/celltypist/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/cross_check_genes.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/cross_check_genes.py
new file mode 100644
index 0000000..b4d5779
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/cross_check_genes.py
@@ -0,0 +1,26 @@
+from typing import List
+
+
+def cross_check_genes(
+    query_genes: List[str], reference_genes: List[str], min_gene_overlap: int = 100
+) -> List[str]:
+    """Cross check the overlap between two lists of genes
+
+    Parameters
+    ----------
+    query_genes : List[str]
+        List of gene names
+    reference_genes : List[str]
+       List of gene names
+
+    Returns
+    -------
+    List[str]
+        List of overlapping genes
+    """
+    common_ens_ids = list(set(reference_genes).intersection(set(query_genes)))
+    assert len(common_ens_ids) >= min_gene_overlap, (
+        f"The intersection of genes between the query and reference dataset is too small, expected at least {min_gene_overlap}."
+    )
+
+    return common_ens_ids
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/is_lognormalized.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/is_lognormalized.py
new file mode 100644
index 0000000..74226c5
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/is_lognormalized.py
@@ -0,0 +1,14 @@
+import numpy as np
+
+
+def is_lognormalized(count_matrix, target_sum=None):
+    if not target_sum:
+        # If no target sum unknown, use first observation's sum as reference
+        target_sum = np.asarray(np.expm1(count_matrix).sum(axis=1)).ravel()[0]
+    exp_row_sums = np.expm1(count_matrix).sum(axis=1)
+    # Check if exponentiated row sums match target (log-normalized data)
+    # and don't overflow (exponentiated row sums become infinity for raw data)
+    if np.isfinite(exp_row_sums).all() and np.allclose(exp_row_sums, target_sum):
+        return True
+
+    return False
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/main.nf
new file mode 100644
index 0000000..c0b6ae1
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/main.nf
@@ -0,0 +1,4328 @@
+// celltypist v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Jakub Majercik (author)
+//  * Weiwei Schultz (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "celltypist",
+  "namespace" : "annotate",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Jakub Majercik",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "links" : {
+          "email" : "jakub@data-intuitive.com",
+          "github" : "jakubmajercik",
+          "linkedin" : "jakubmajercik"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Bioinformatics Engineer"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "description" : "Input dataset (query) arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "The input (query) data to be labeled. Should be a .h5mu file.",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_layer",
+          "description" : "The layer in the input data containing counts that are lognormalized to 10000, .X is not to be used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_var_gene_names",
+          "description" : "The name of the adata var column in the input data containing gene names; when no gene_name_layer is provided, the var index will be used.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--input_reference_gene_overlap",
+          "description" : "The minimum number of genes present in both the reference and query datasets.\n",
+          "default" : [
+            100
+          ],
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--sanitize_ensembl_ids",
+          "description" : "Whether to sanitize ensembl ids by removing version numbers.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Reference",
+      "description" : "Arguments related to the reference dataset.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--reference",
+          "description" : "The reference data to train the CellTypist classifiers on. Only required if a pre-trained --model is not provided.",
+          "example" : [
+            "reference.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_layer",
+          "description" : "The layer in the reference data containing counts that are lognormalized to 10000, if .X is not to be used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_obs_target",
+          "description" : "The name of the adata obs column in the reference data containing cell type annotations.",
+          "default" : [
+            "cell_ontology_class"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_var_input",
+          "description" : ".var column containing highly variable genes. By default, do not subset genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_var_gene_names",
+          "description" : "The name of the adata var column in the reference data containing gene names; when no gene_name_layer is provided, the var index will be used.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Model arguments",
+      "description" : "Model arguments.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--model",
+          "description" : "Pretrained model in pkl format. If not provided, the model will be trained on the reference data and --reference should be provided.",
+          "example" : [
+            "pretrained_model.pkl"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--feature_selection",
+          "description" : "Whether to perform feature selection.",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--majority_voting",
+          "description" : "Whether to refine the predicted labels by running the majority voting classifier after over-clustering.",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--C",
+          "description" : "Inverse of regularization strength in logistic regression.",
+          "default" : [
+            1.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_iter",
+          "description" : "Maximum number of iterations before reaching the minimum of the cost function.",
+          "default" : [
+            1000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--use_SGD",
+          "description" : "Whether to use the stochastic gradient descent algorithm.",
+          "direction" : "input"
+        },
+        {
+          "type" : "double",
+          "name" : "--min_prop",
+          "description" : "\\"For the dominant cell type within a subcluster, the minimum proportion of cells required to \nsupport naming of the subcluster by this cell type. Ignored if majority_voting is set to False. \nSubcluster that fails to pass this proportion threshold will be assigned 'Heterogeneous'.\\"\n",
+          "default" : [
+            0.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "description" : "Output arguments.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Output h5mu file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_predictions",
+          "description" : "In which `.obs` slots to store the predicted information.\n",
+          "default" : [
+            "celltypist_pred"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_probability",
+          "description" : "In which `.obs` slots to store the probability of the predictions.\n",
+          "default" : [
+            "celltypist_probability"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/cross_check_genes.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/subset_vars.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/set_var_index.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/is_lognormalized.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Automated cell type annotation tool for scRNA-seq datasets on the basis of logistic regression classifiers optimised by the stochastic gradient descent algorithm.",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/annotation_test_data/"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "highcpu",
+          "highmem",
+          "highdisk",
+          "gpu"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "nvcr.io/nvidia/pytorch:25.11-py3",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "scanpy~=1.11.4"
+          ],
+          "upgrade" : true
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "celltypist==1.7.1"
+          ],
+          "upgrade" : true
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/annotate/celltypist/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/annotate/celltypist",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+import celltypist
+import mudata as mu
+import anndata as ad
+import pandas as pd
+from torch.cuda import is_available as cuda_is_available
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_layer': $( if [ ! -z ${VIASH_PAR_INPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_INPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_var_gene_names': $( if [ ! -z ${VIASH_PAR_INPUT_VAR_GENE_NAMES+x} ]; then echo "r'${VIASH_PAR_INPUT_VAR_GENE_NAMES//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_reference_gene_overlap': $( if [ ! -z ${VIASH_PAR_INPUT_REFERENCE_GENE_OVERLAP+x} ]; then echo "int(r'${VIASH_PAR_INPUT_REFERENCE_GENE_OVERLAP//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'sanitize_ensembl_ids': $( if [ ! -z ${VIASH_PAR_SANITIZE_ENSEMBL_IDS+x} ]; then echo "r'${VIASH_PAR_SANITIZE_ENSEMBL_IDS//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'reference': $( if [ ! -z ${VIASH_PAR_REFERENCE+x} ]; then echo "r'${VIASH_PAR_REFERENCE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'reference_layer': $( if [ ! -z ${VIASH_PAR_REFERENCE_LAYER+x} ]; then echo "r'${VIASH_PAR_REFERENCE_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'reference_obs_target': $( if [ ! -z ${VIASH_PAR_REFERENCE_OBS_TARGET+x} ]; then echo "r'${VIASH_PAR_REFERENCE_OBS_TARGET//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'reference_var_input': $( if [ ! -z ${VIASH_PAR_REFERENCE_VAR_INPUT+x} ]; then echo "r'${VIASH_PAR_REFERENCE_VAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'reference_var_gene_names': $( if [ ! -z ${VIASH_PAR_REFERENCE_VAR_GENE_NAMES+x} ]; then echo "r'${VIASH_PAR_REFERENCE_VAR_GENE_NAMES//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'model': $( if [ ! -z ${VIASH_PAR_MODEL+x} ]; then echo "r'${VIASH_PAR_MODEL//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'feature_selection': $( if [ ! -z ${VIASH_PAR_FEATURE_SELECTION+x} ]; then echo "r'${VIASH_PAR_FEATURE_SELECTION//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'majority_voting': $( if [ ! -z ${VIASH_PAR_MAJORITY_VOTING+x} ]; then echo "r'${VIASH_PAR_MAJORITY_VOTING//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'C': $( if [ ! -z ${VIASH_PAR_C+x} ]; then echo "float(r'${VIASH_PAR_C//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'max_iter': $( if [ ! -z ${VIASH_PAR_MAX_ITER+x} ]; then echo "int(r'${VIASH_PAR_MAX_ITER//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'use_SGD': $( if [ ! -z ${VIASH_PAR_USE_SGD+x} ]; then echo "r'${VIASH_PAR_USE_SGD//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'min_prop': $( if [ ! -z ${VIASH_PAR_MIN_PROP+x} ]; then echo "float(r'${VIASH_PAR_MIN_PROP//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_obs_predictions': $( if [ ! -z ${VIASH_PAR_OUTPUT_OBS_PREDICTIONS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_OBS_PREDICTIONS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_obs_probability': $( if [ ! -z ${VIASH_PAR_OUTPUT_OBS_PROBABILITY+x} ]; then echo "r'${VIASH_PAR_OUTPUT_OBS_PROBABILITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from cross_check_genes import cross_check_genes
+from set_var_index import set_var_index
+from subset_vars import subset_vars
+from is_lognormalized import is_lognormalized
+
+logger = setup_logger()
+use_gpu = cuda_is_available()
+logger.info("GPU enabled? %s", use_gpu)
+
+
+def main(par):
+    if (not par["model"] and not par["reference"]) or (
+        par["model"] and par["reference"]
+    ):
+        raise ValueError(
+            "Make sure to provide either 'model' or 'reference', but not both."
+        )
+
+    input_mudata = mu.read_h5mu(par["input"])
+    input_adata = input_mudata.mod[par["modality"]]
+    input_modality = input_adata.copy()
+
+    # Provide correct format of query data for celltypist annotation
+    ## Sanitize gene names and set as index
+    input_modality = set_var_index(
+        input_modality, par["input_var_gene_names"], par["sanitize_ensembl_ids"]
+    )
+
+    ## Fetch lognormalized counts
+    lognorm_counts = (
+        input_modality.layers[par["input_layer"]].copy()
+        if par["input_layer"]
+        else input_modality.X.copy()
+    )
+
+    if not is_lognormalized(lognorm_counts, target_sum=10000):
+        raise ValueError(
+            "Invalid expression matrix, expect input layer to contain log1p normalized expression to 10000 counts per cell."
+        )
+
+    ## Create AnnData object
+    input_modality = ad.AnnData(
+        X=lognorm_counts, var=pd.DataFrame(index=input_modality.var.index)
+    )
+
+    if par["model"]:
+        logger.info("Loading CellTypist model")
+        model = celltypist.models.Model.load(par["model"])
+        cross_check_genes(
+            input_modality.var.index,
+            model.features,
+            min_gene_overlap=par["input_reference_gene_overlap"],
+        )
+
+    elif par["reference"]:
+        reference_modality = mu.read_h5mu(par["reference"]).mod[par["modality"]]
+
+        # Check expression before subsetting to HVG
+        if not is_lognormalized(lognorm_counts, target_sum=10000):
+            raise ValueError(
+                "Invalid expression matrix, expect reference layer to contain log1p normalized expression to 10000 counts per cell."
+            )
+
+        # subset to HVG if required
+        if par["reference_var_input"]:
+            reference_modality = subset_vars(
+                reference_modality, par["reference_var_input"]
+            )
+
+        # Set var names to the desired gene name format (gene symbol, ensembl id, etc.)
+        # CellTypist requires query gene names to be in index
+        reference_modality = set_var_index(
+            reference_modality,
+            par["reference_var_gene_names"],
+            par["sanitize_ensembl_ids"],
+        )
+
+        # Ensure enough overlap between genes in query and reference
+        cross_check_genes(
+            input_modality.var.index,
+            reference_modality.var.index,
+            min_gene_overlap=par["input_reference_gene_overlap"],
+        )
+
+        reference_matrix = (
+            reference_modality.layers[par["reference_layer"]]
+            if par["reference_layer"]
+            else reference_modality.X
+        )
+
+        labels = reference_modality.obs[par["reference_obs_target"]]
+
+        logger.info("Training CellTypist model on reference")
+        model = celltypist.train(
+            reference_matrix,
+            labels=labels,
+            genes=reference_modality.var.index.astype(
+                str
+            ),  # ensure string type, fails if categorical when making var indices unique
+            C=par["C"],
+            max_iter=par["max_iter"],
+            use_SGD=par["use_SGD"],
+            feature_selection=par["feature_selection"],
+            check_expression=False,  # if True, throws an error if lognormalized data are subset for HVG,
+            use_GPU=use_gpu,
+        )
+
+    logger.info("Predicting CellTypist annotations")
+    # Make sure .var index is string dtype, fails if categorical when making var indices unique
+    input_modality.var.index = input_modality.var.index.astype(str)
+    predictions = celltypist.annotate(
+        input_modality, model, majority_voting=par["majority_voting"], use_GPU=use_gpu
+    )
+
+    input_adata.obs[par["output_obs_predictions"]] = predictions.predicted_labels[
+        "predicted_labels"
+    ].values
+    input_adata.obs[par["output_obs_probability"]] = predictions.probability_matrix.max(
+        axis=1
+    ).values
+
+    # copy observations back to input data (with full set of features)
+    input_mudata.write_h5mu(par["output"], compression=par["output_compression"])
+
+
+if __name__ == "__main__":
+    main(par)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/annotate/celltypist",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "highcpu",
+    "highmem",
+    "highdisk",
+    "gpu"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/nextflow.config
new file mode 100644
index 0000000..e38e355
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'annotate/celltypist'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Automated cell type annotation tool for scRNA-seq datasets on the basis of logistic regression classifiers optimised by the stochastic gradient descent algorithm.'
+  author = 'Jakub Majercik, Weiwei Schultz'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/nextflow_schema.json
new file mode 100644
index 0000000..08e318c
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/nextflow_schema.json
@@ -0,0 +1,197 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "celltypist",
+  "description": "Automated cell type annotation tool for scRNA-seq datasets on the basis of logistic regression classifiers optimised by the stochastic gradient descent algorithm.",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "Input dataset (query) arguments",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "The input (query) data to be labeled",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality to process.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "input_layer": {
+          "type": "string",
+          "description": "The layer in the input data containing counts that are lognormalized to 10000, .X is not to be used.",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "input_var_gene_names": {
+          "type": "string",
+          "description": "The name of the adata var column in the input data containing gene names; when no gene_name_layer is provided, the var index will be used.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "input_reference_gene_overlap": {
+          "type": "integer",
+          "description": "The minimum number of genes present in both the reference and query datasets.\n",
+          "help_text": "Type: `integer`, multiple: `False`, default: `100`. ",
+          "default": 100
+        },
+        "sanitize_ensembl_ids": {
+          "type": "boolean",
+          "description": "Whether to sanitize ensembl ids by removing version numbers.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "Output arguments.",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "output_obs_predictions": {
+          "type": "string",
+          "description": "In which `.obs` slots to store the predicted information.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"celltypist_pred\"`. ",
+          "default": "celltypist_pred"
+        },
+        "output_obs_probability": {
+          "type": "string",
+          "description": "In which `.obs` slots to store the probability of the predictions.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"celltypist_probability\"`. ",
+          "default": "celltypist_probability"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "reference": {
+      "title": "Reference",
+      "type": "object",
+      "description": "Arguments related to the reference dataset.",
+      "properties": {
+        "reference": {
+          "type": "string",
+          "format": "path",
+          "description": "The reference data to train the CellTypist classifiers on",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"reference.h5mu\"`. "
+        },
+        "reference_layer": {
+          "type": "string",
+          "description": "The layer in the reference data containing counts that are lognormalized to 10000, if .X is not to be used.",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "reference_obs_target": {
+          "type": "string",
+          "description": "The name of the adata obs column in the reference data containing cell type annotations.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"cell_ontology_class\"`. ",
+          "default": "cell_ontology_class"
+        },
+        "reference_var_input": {
+          "type": "string",
+          "description": ".var column containing highly variable genes",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "reference_var_gene_names": {
+          "type": "string",
+          "description": "The name of the adata var column in the reference data containing gene names; when no gene_name_layer is provided, the var index will be used.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "model arguments": {
+      "title": "Model arguments",
+      "type": "object",
+      "description": "Model arguments.",
+      "properties": {
+        "model": {
+          "type": "string",
+          "format": "path",
+          "description": "Pretrained model in pkl format",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"pretrained_model.pkl\"`. "
+        },
+        "feature_selection": {
+          "type": "boolean",
+          "description": "Whether to perform feature selection.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "majority_voting": {
+          "type": "boolean",
+          "description": "Whether to refine the predicted labels by running the majority voting classifier after over-clustering.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "C": {
+          "type": "number",
+          "description": "Inverse of regularization strength in logistic regression.",
+          "help_text": "Type: `double`, multiple: `False`, default: `1.0`. ",
+          "default": 1.0
+        },
+        "max_iter": {
+          "type": "integer",
+          "description": "Maximum number of iterations before reaching the minimum of the cost function.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `1000`. ",
+          "default": 1000
+        },
+        "use_SGD": {
+          "type": "boolean",
+          "description": "Whether to use the stochastic gradient descent algorithm.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "min_prop": {
+          "type": "number",
+          "description": "\"For the dominant cell type within a subcluster, the minimum proportion of cells required to \nsupport naming of the subcluster by this cell type",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.0`. ",
+          "default": 0.0
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/reference"
+    },
+    {
+      "$ref": "#/$defs/model arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/set_var_index.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/set_var_index.py
new file mode 100644
index 0000000..1dd7e0d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/set_var_index.py
@@ -0,0 +1,68 @@
+import anndata as ad
+
+
+def strip_version_number(gene_series: list[str]) -> list[str]:
+    """Sanitize ensemble ID's by removing version numbers.
+
+    Parameters
+    ----------
+    gene_series : list[str]
+        List of ensemble ID's to sanitize.
+
+    Returns
+    -------
+    list[str]
+        List of sanitized ensemble ID's.
+    """
+
+    # Convert to string type to handle Categorical series
+    gene_series = gene_series.astype(str)
+
+    # Pattern matches Ensembl IDs: starts with ENS, followed by any characters,
+    # then an eleven digit number, optionally followed by .version_number
+    ensembl_pattern = r"^(ENS.*\d{11})(?:\.\d+)?$"
+    ensembl_mask = gene_series.str.match(ensembl_pattern)
+
+    sanitized = gene_series.where(
+        ~ensembl_mask, gene_series.str.extract(ensembl_pattern)[0]
+    )
+
+    return sanitized
+
+
+def set_var_index(
+    adata: ad.AnnData, var_name: str | None = None, sanitize_ensembl_ids: bool = True
+) -> ad.AnnData:
+    """Sanitize gene names (optional) and set the index of the .var DataFrame.
+
+    Parameters
+    ----------
+    adata : AnnData
+        Annotated data object
+    var_name : str | None
+        Name of the column in `adata.var` that contains the gene names, if None, the existing index will be sanitized but not replaced.
+    sanitize_ensembl_ids : bool
+        Whether to sanitize gene names by removing version numbers.
+
+    Returns
+    -------
+    AnnData
+        Copy of `adata` with optionally sanitized and replaced index
+    """
+    gene_names = adata.var[var_name] if var_name else adata.var.index.to_series()
+
+    if sanitize_ensembl_ids:
+        ori_gene_names = len(gene_names)
+        gene_names = strip_version_number(gene_names)
+        sanitized_gene_names = len(set(gene_names))
+
+        assert ori_gene_names == sanitized_gene_names, (
+            "Sanitizing gene names resulted in duplicated gene names.\n"
+            "Please ensure unique gene names before proceeding.\n"
+            "Please make sure --var_gene_names contains ensembl IDs (not gene symbols) "
+            "when --sanitize_ensembl_ids is set to True."
+        )
+
+    adata.var.index = gene_names.astype(str)
+
+    return adata
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/subset_vars.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/subset_vars.py
new file mode 100644
index 0000000..6003513
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/subset_vars.py
@@ -0,0 +1,31 @@
+def subset_vars(adata, subset_col):
+    """Subset AnnData object on highly variable genes
+
+    Parameters
+    ----------
+    adata : AnnData
+        Annotated data object
+    subset_col : str
+        Name of the boolean column in `adata.var` that contains the information if features should be used or not
+
+    Returns
+    -------
+    AnnData
+        Copy of `adata` with subsetted features
+    """
+    if subset_col not in adata.var.columns:
+        raise ValueError(
+            f"Requested to use .var column '{subset_col}' as a selection of genes, but the column is not available."
+        )
+
+    if adata.var[subset_col].dtype == "boolean":
+        assert adata.var[subset_col].isna().sum() == 0, (
+            f"The .var column `{subset_col}` contains NaN values. Can not subset data."
+        )
+        adata.var[subset_col] = adata.var[subset_col].astype("bool")
+
+    assert adata.var[subset_col].dtype == "bool", (
+        f"Expected dtype of .var column '{subset_col}' to be `bool`, but found {adata.var[subset_col].dtype}. Can not subset data."
+    )
+
+    return adata[:, adata.var[subset_col]].copy()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/.config.vsh.yaml
new file mode 100644
index 0000000..5035018
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/.config.vsh.yaml
@@ -0,0 +1,490 @@
+name: "scanvi"
+namespace: "annotate"
+version: "v4.0.4"
+authors:
+- name: "Dorien Roosen"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Jakub Majercik"
+  roles:
+  - "author"
+  info:
+    role: "Contributor"
+    links:
+      email: "jakub@data-intuitive.com"
+      github: "jakubmajercik"
+      linkedin: "jakubmajercik"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Bioinformatics Engineer"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file. Note that this needs to be the exact same dataset\
+      \ as the --scvi_model was trained on."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_layer"
+    description: "Input layer to use. If None, X is used"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_input"
+    description: ".var column containing highly variable genes that were used to train\
+      \ the scVi model. By default, do not subset genes."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_gene_names"
+    description: ".var column containing gene names. By default, use the index."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_labels"
+    description: ".obs field containing the labels"
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--unlabeled_category"
+    description: "Value in the --obs_labels field that indicates unlabeled observations\n"
+    info: null
+    default:
+    - "Unknown"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--sanitize_ensembl_ids"
+    description: "Whether to sanitize ensembl ids by removing version numbers."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "scVI Model"
+  arguments:
+  - type: "file"
+    name: "--scvi_model"
+    description: "Pretrained SCVI reference model to initialize the SCANVI model with."
+    info: null
+    example:
+    - "scvi_model.pt"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_model"
+    description: "Folder where the state of the trained model will be saved to."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_output"
+    description: "In which .obsm slot to store the resulting integrated embedding."
+    info: null
+    default:
+    - "X_scanvi_integrated"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_output_predictions"
+    description: "In which .obs slot to store the predicted labels."
+    info: null
+    default:
+    - "scanvi_pred"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_output_probabilities"
+    description: "In which. obs slot to store the probabilities of the predicted labels."
+    info: null
+    default:
+    - "scanvi_proba"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "scANVI training arguments"
+  arguments:
+  - type: "boolean"
+    name: "--early_stopping"
+    description: "Whether to perform early stopping with respect to the validation\
+      \ set."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--early_stopping_monitor"
+    description: "Metric logged during validation set epoch."
+    info: null
+    default:
+    - "elbo_validation"
+    required: false
+    choices:
+    - "elbo_validation"
+    - "reconstruction_loss_validation"
+    - "kl_local_validation"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--early_stopping_patience"
+    description: "Number of validation epochs with no improvement after which training\
+      \ will be stopped."
+    info: null
+    default:
+    - 45
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--early_stopping_min_delta"
+    description: "Minimum change in the monitored quantity to qualify as an improvement,\
+      \ i.e. an absolute change of less than min_delta, will count as no improvement."
+    info: null
+    default:
+    - 0.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_epochs"
+    description: "Number of passes through the dataset, defaults to (20000 / number\
+      \ of cells) * 400 or 400; whichever is smallest."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--reduce_lr_on_plateau"
+    description: "Whether to monitor validation loss and reduce learning rate when\
+      \ validation set `lr_scheduler_metric` plateaus."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--lr_factor"
+    description: "Factor to reduce learning rate."
+    info: null
+    default:
+    - 0.6
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--lr_patience"
+    description: "Number of epochs with no improvement after which learning rate will\
+      \ be reduced."
+    info: null
+    default:
+    - 30.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "subset_vars.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "set_var_index.py"
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "scANVI () is a semi-supervised model for single-cell transcriptomics\
+  \ data. scANVI is an scVI extension that can leverage the cell type knowledge for\
+  \ a subset of the cells present in the data sets to infer the states of the rest\
+  \ of the cells.\nThis component will instantiate a scANVI model from a pre-trained\
+  \ scVI model, integrate the data and perform label prediction.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "scvi_model"
+- type: "file"
+  path: "TS_Blood_filtered.h5mu"
+- type: "file"
+  path: "pbmc_1k_protein_v3_mms.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "midcpu"
+    - "midmem"
+    - "gpu"
+    - "highdisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "pytorch/pytorch:2.9.1-cuda12.8-cudnn9-runtime"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  - type: "python"
+    user: false
+    packages:
+    - "scvi-tools[cuda]~=1.4.1"
+    upgrade: true
+  test_setup:
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/annotate/scanvi/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/annotate/scanvi"
+  executable: "target/nextflow/annotate/scanvi/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/main.nf
new file mode 100644
index 0000000..feefc95
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/main.nf
@@ -0,0 +1,4298 @@
+// scanvi v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dorien Roosen (maintainer)
+//  * Jakub Majercik (author)
+//  * Weiwei Schultz (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "scanvi",
+  "namespace" : "annotate",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Jakub Majercik",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "links" : {
+          "email" : "jakub@data-intuitive.com",
+          "github" : "jakubmajercik",
+          "linkedin" : "jakubmajercik"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Bioinformatics Engineer"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file. Note that this needs to be the exact same dataset as the --scvi_model was trained on.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_layer",
+          "description" : "Input layer to use. If None, X is used",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_input",
+          "description" : ".var column containing highly variable genes that were used to train the scVi model. By default, do not subset genes.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_gene_names",
+          "description" : ".var column containing gene names. By default, use the index.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_labels",
+          "description" : ".obs field containing the labels",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--unlabeled_category",
+          "description" : "Value in the --obs_labels field that indicates unlabeled observations\n",
+          "default" : [
+            "Unknown"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--sanitize_ensembl_ids",
+          "description" : "Whether to sanitize ensembl ids by removing version numbers.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "scVI Model",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--scvi_model",
+          "description" : "Pretrained SCVI reference model to initialize the SCANVI model with.",
+          "example" : [
+            "scvi_model.pt"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_model",
+          "description" : "Folder where the state of the trained model will be saved to.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_output",
+          "description" : "In which .obsm slot to store the resulting integrated embedding.",
+          "default" : [
+            "X_scanvi_integrated"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_output_predictions",
+          "description" : "In which .obs slot to store the predicted labels.",
+          "default" : [
+            "scanvi_pred"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_output_probabilities",
+          "description" : "In which. obs slot to store the probabilities of the predicted labels.",
+          "default" : [
+            "scanvi_proba"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "scANVI training arguments",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--early_stopping",
+          "description" : "Whether to perform early stopping with respect to the validation set.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--early_stopping_monitor",
+          "description" : "Metric logged during validation set epoch.",
+          "default" : [
+            "elbo_validation"
+          ],
+          "required" : false,
+          "choices" : [
+            "elbo_validation",
+            "reconstruction_loss_validation",
+            "kl_local_validation"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--early_stopping_patience",
+          "description" : "Number of validation epochs with no improvement after which training will be stopped.",
+          "default" : [
+            45
+          ],
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--early_stopping_min_delta",
+          "description" : "Minimum change in the monitored quantity to qualify as an improvement, i.e. an absolute change of less than min_delta, will count as no improvement.",
+          "default" : [
+            0.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_epochs",
+          "description" : "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--reduce_lr_on_plateau",
+          "description" : "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--lr_factor",
+          "description" : "Factor to reduce learning rate.",
+          "default" : [
+            0.6
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--lr_patience",
+          "description" : "Number of epochs with no improvement after which learning rate will be reduced.",
+          "default" : [
+            30.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/subset_vars.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/set_var_index.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "scANVI () is a semi-supervised model for single-cell transcriptomics data. scANVI is an scVI extension that can leverage the cell type knowledge for a subset of the cells present in the data sets to infer the states of the rest of the cells.\nThis component will instantiate a scANVI model from a pre-trained scVI model, integrate the data and perform label prediction.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/annotation_test_data/scvi_model/"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/annotation_test_data/TS_Blood_filtered.h5mu"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_mms.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "midcpu",
+          "midmem",
+          "gpu",
+          "highdisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "pytorch/pytorch:2.9.1-cuda12.8-cudnn9-runtime",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "scvi-tools[cuda]~=1.4.1"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/annotate/scanvi/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/annotate/scanvi",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import mudata
+import scvi
+import numpy as np
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_layer': $( if [ ! -z ${VIASH_PAR_INPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_INPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_input': $( if [ ! -z ${VIASH_PAR_VAR_INPUT+x} ]; then echo "r'${VIASH_PAR_VAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_gene_names': $( if [ ! -z ${VIASH_PAR_VAR_GENE_NAMES+x} ]; then echo "r'${VIASH_PAR_VAR_GENE_NAMES//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_labels': $( if [ ! -z ${VIASH_PAR_OBS_LABELS+x} ]; then echo "r'${VIASH_PAR_OBS_LABELS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'unlabeled_category': $( if [ ! -z ${VIASH_PAR_UNLABELED_CATEGORY+x} ]; then echo "r'${VIASH_PAR_UNLABELED_CATEGORY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'sanitize_ensembl_ids': $( if [ ! -z ${VIASH_PAR_SANITIZE_ENSEMBL_IDS+x} ]; then echo "r'${VIASH_PAR_SANITIZE_ENSEMBL_IDS//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'scvi_model': $( if [ ! -z ${VIASH_PAR_SCVI_MODEL+x} ]; then echo "r'${VIASH_PAR_SCVI_MODEL//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_model': $( if [ ! -z ${VIASH_PAR_OUTPUT_MODEL+x} ]; then echo "r'${VIASH_PAR_OUTPUT_MODEL//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_output': $( if [ ! -z ${VIASH_PAR_OBSM_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OBSM_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_output_predictions': $( if [ ! -z ${VIASH_PAR_OBS_OUTPUT_PREDICTIONS+x} ]; then echo "r'${VIASH_PAR_OBS_OUTPUT_PREDICTIONS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_output_probabilities': $( if [ ! -z ${VIASH_PAR_OBS_OUTPUT_PROBABILITIES+x} ]; then echo "r'${VIASH_PAR_OBS_OUTPUT_PROBABILITIES//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'early_stopping': $( if [ ! -z ${VIASH_PAR_EARLY_STOPPING+x} ]; then echo "r'${VIASH_PAR_EARLY_STOPPING//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'early_stopping_monitor': $( if [ ! -z ${VIASH_PAR_EARLY_STOPPING_MONITOR+x} ]; then echo "r'${VIASH_PAR_EARLY_STOPPING_MONITOR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'early_stopping_patience': $( if [ ! -z ${VIASH_PAR_EARLY_STOPPING_PATIENCE+x} ]; then echo "int(r'${VIASH_PAR_EARLY_STOPPING_PATIENCE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'early_stopping_min_delta': $( if [ ! -z ${VIASH_PAR_EARLY_STOPPING_MIN_DELTA+x} ]; then echo "float(r'${VIASH_PAR_EARLY_STOPPING_MIN_DELTA//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'max_epochs': $( if [ ! -z ${VIASH_PAR_MAX_EPOCHS+x} ]; then echo "int(r'${VIASH_PAR_MAX_EPOCHS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'reduce_lr_on_plateau': $( if [ ! -z ${VIASH_PAR_REDUCE_LR_ON_PLATEAU+x} ]; then echo "r'${VIASH_PAR_REDUCE_LR_ON_PLATEAU//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'lr_factor': $( if [ ! -z ${VIASH_PAR_LR_FACTOR+x} ]; then echo "float(r'${VIASH_PAR_LR_FACTOR//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'lr_patience': $( if [ ! -z ${VIASH_PAR_LR_PATIENCE+x} ]; then echo "float(r'${VIASH_PAR_LR_PATIENCE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+import sys
+
+sys.path.append(meta["resources_dir"])
+
+from subset_vars import subset_vars
+from set_var_index import set_var_index
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+
+def main():
+    logger.info("Reading input data...")
+    adata = mudata.read_h5ad(par["input"].strip(), mod=par["modality"])
+
+    if par["var_input"]:
+        # Subset to HVG
+        adata_subset = subset_vars(adata, subset_col=par["var_input"]).copy()
+    else:
+        adata_subset = adata.copy()
+
+    # Sanitize gene names and set as index of the AnnData object
+    adata_subset = set_var_index(
+        adata_subset, par["var_gene_names"], par["sanitize_ensembl_ids"]
+    )
+
+    logger.info(f"Loading pre-trained scVI model from {par['scvi_model']}")
+    scvi_model = scvi.model.SCVI.load(
+        par["scvi_model"],
+        adata_subset,
+        accelerator="auto",
+        device="auto",
+    )
+
+    logger.info("Instantiating scANVI model from scVI model...")
+    vae_uns = scvi.model.SCANVI.from_scvi_model(
+        scvi_model,
+        unlabeled_category=par["unlabeled_category"],
+        labels_key=par["obs_labels"],
+        adata=adata_subset,
+    )
+
+    plan_kwargs = {
+        "reduce_lr_on_plateau": par["reduce_lr_on_plateau"],
+        "lr_patience": par["lr_patience"],
+        "lr_factor": par["lr_factor"],
+    }
+
+    logger.info("Training scANVI model...")
+    # Train the model
+    vae_uns.train(
+        max_epochs=par["max_epochs"],
+        early_stopping=par["early_stopping"],
+        early_stopping_monitor=par["early_stopping_monitor"],
+        early_stopping_patience=par["early_stopping_patience"],
+        early_stopping_min_delta=par["early_stopping_min_delta"],
+        plan_kwargs=plan_kwargs,
+        check_val_every_n_epoch=1,
+        accelerator="auto",
+    )
+    # Note: train_size=1.0 should give better results, but then can't do early_stopping on validation set
+
+    logger.info("Performing scANVI integration...")
+    # Get the latent output
+    adata.obsm[par["obsm_output"]] = vae_uns.get_latent_representation()
+
+    logger.info("Performing scANVI prediction...")
+    adata.obs[par["obs_output_predictions"]] = vae_uns.predict()
+    adata.obs[par["obs_output_probabilities"]] = np.max(
+        vae_uns.predict(soft=True), axis=1
+    )
+
+    logger.info("Writing output data...")
+    write_h5ad_to_h5mu_with_compression(
+        par["output"], par["input"], par["modality"], adata, par["output_compression"]
+    )
+    if par["output_model"]:
+        vae_uns.save(par["output_model"], overwrite=True)
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/annotate/scanvi",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "midcpu",
+    "midmem",
+    "gpu",
+    "highdisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/nextflow.config
new file mode 100644
index 0000000..960cff0
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'annotate/scanvi'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'scANVI () is a semi-supervised model for single-cell transcriptomics data. scANVI is an scVI extension that can leverage the cell type knowledge for a subset of the cells present in the data sets to infer the states of the rest of the cells.\nThis component will instantiate a scANVI model from a pre-trained scVI model, integrate the data and perform label prediction.\n'
+  author = 'Dorien Roosen, Jakub Majercik, Weiwei Schultz'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/nextflow_schema.json
new file mode 100644
index 0000000..5393d04
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/nextflow_schema.json
@@ -0,0 +1,209 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "scanvi",
+  "description": "scANVI () is a semi-supervised model for single-cell transcriptomics data. scANVI is an scVI extension that can leverage the cell type knowledge for a subset of the cells present in the data sets to infer the states of the rest of the cells.\nThis component will instantiate a scANVI model from a pre-trained scVI model, integrate the data and perform label prediction.\n",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "input_layer": {
+          "type": "string",
+          "description": "Input layer to use",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "var_input": {
+          "type": "string",
+          "description": ".var column containing highly variable genes that were used to train the scVi model",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "var_gene_names": {
+          "type": "string",
+          "description": ".var column containing gene names",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_labels": {
+          "type": "string",
+          "description": ".obs field containing the labels",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        },
+        "unlabeled_category": {
+          "type": "string",
+          "description": "Value in the --obs_labels field that indicates unlabeled observations\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"Unknown\"`. ",
+          "default": "Unknown"
+        },
+        "sanitize_ensembl_ids": {
+          "type": "boolean",
+          "description": "Whether to sanitize ensembl ids by removing version numbers.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output\"`, direction: `output`. ",
+          "default": "$id.$key.output"
+        },
+        "output_model": {
+          "type": "string",
+          "format": "path",
+          "description": "Folder where the state of the trained model will be saved to.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output_model\"`, direction: `output`. ",
+          "default": "$id.$key.output_model"
+        },
+        "obsm_output": {
+          "type": "string",
+          "description": "In which .obsm slot to store the resulting integrated embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_scanvi_integrated\"`. ",
+          "default": "X_scanvi_integrated"
+        },
+        "obs_output_predictions": {
+          "type": "string",
+          "description": "In which .obs slot to store the predicted labels.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scanvi_pred\"`. ",
+          "default": "scanvi_pred"
+        },
+        "obs_output_probabilities": {
+          "type": "string",
+          "description": "In which",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scanvi_proba\"`. ",
+          "default": "scanvi_proba"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "scvi model": {
+      "title": "scVI Model",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "scvi_model": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Pretrained SCVI reference model to initialize the SCANVI model with.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"scvi_model.pt\"`. "
+        }
+      }
+    },
+    "scanvi training arguments": {
+      "title": "scANVI training arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "early_stopping": {
+          "type": "boolean",
+          "description": "Whether to perform early stopping with respect to the validation set.",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        },
+        "early_stopping_monitor": {
+          "type": "string",
+          "description": "Metric logged during validation set epoch.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"elbo_validation\"`, choices: ``elbo_validation`, `reconstruction_loss_validation`, `kl_local_validation``. ",
+          "enum": [
+            "elbo_validation",
+            "reconstruction_loss_validation",
+            "kl_local_validation"
+          ],
+          "default": "elbo_validation"
+        },
+        "early_stopping_patience": {
+          "type": "integer",
+          "description": "Number of validation epochs with no improvement after which training will be stopped.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `45`. ",
+          "default": 45
+        },
+        "early_stopping_min_delta": {
+          "type": "number",
+          "description": "Minimum change in the monitored quantity to qualify as an improvement, i.e",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.0`. ",
+          "default": 0.0
+        },
+        "max_epochs": {
+          "type": "integer",
+          "description": "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest.",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "reduce_lr_on_plateau": {
+          "type": "boolean",
+          "description": "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "lr_factor": {
+          "type": "number",
+          "description": "Factor to reduce learning rate.",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.6`. ",
+          "default": 0.6
+        },
+        "lr_patience": {
+          "type": "number",
+          "description": "Number of epochs with no improvement after which learning rate will be reduced.",
+          "help_text": "Type: `double`, multiple: `False`, default: `30.0`. ",
+          "default": 30.0
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/scvi model"
+    },
+    {
+      "$ref": "#/$defs/scanvi training arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/set_var_index.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/set_var_index.py
new file mode 100644
index 0000000..1dd7e0d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/set_var_index.py
@@ -0,0 +1,68 @@
+import anndata as ad
+
+
+def strip_version_number(gene_series: list[str]) -> list[str]:
+    """Sanitize ensemble ID's by removing version numbers.
+
+    Parameters
+    ----------
+    gene_series : list[str]
+        List of ensemble ID's to sanitize.
+
+    Returns
+    -------
+    list[str]
+        List of sanitized ensemble ID's.
+    """
+
+    # Convert to string type to handle Categorical series
+    gene_series = gene_series.astype(str)
+
+    # Pattern matches Ensembl IDs: starts with ENS, followed by any characters,
+    # then an eleven digit number, optionally followed by .version_number
+    ensembl_pattern = r"^(ENS.*\d{11})(?:\.\d+)?$"
+    ensembl_mask = gene_series.str.match(ensembl_pattern)
+
+    sanitized = gene_series.where(
+        ~ensembl_mask, gene_series.str.extract(ensembl_pattern)[0]
+    )
+
+    return sanitized
+
+
+def set_var_index(
+    adata: ad.AnnData, var_name: str | None = None, sanitize_ensembl_ids: bool = True
+) -> ad.AnnData:
+    """Sanitize gene names (optional) and set the index of the .var DataFrame.
+
+    Parameters
+    ----------
+    adata : AnnData
+        Annotated data object
+    var_name : str | None
+        Name of the column in `adata.var` that contains the gene names, if None, the existing index will be sanitized but not replaced.
+    sanitize_ensembl_ids : bool
+        Whether to sanitize gene names by removing version numbers.
+
+    Returns
+    -------
+    AnnData
+        Copy of `adata` with optionally sanitized and replaced index
+    """
+    gene_names = adata.var[var_name] if var_name else adata.var.index.to_series()
+
+    if sanitize_ensembl_ids:
+        ori_gene_names = len(gene_names)
+        gene_names = strip_version_number(gene_names)
+        sanitized_gene_names = len(set(gene_names))
+
+        assert ori_gene_names == sanitized_gene_names, (
+            "Sanitizing gene names resulted in duplicated gene names.\n"
+            "Please ensure unique gene names before proceeding.\n"
+            "Please make sure --var_gene_names contains ensembl IDs (not gene symbols) "
+            "when --sanitize_ensembl_ids is set to True."
+        )
+
+    adata.var.index = gene_names.astype(str)
+
+    return adata
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/subset_vars.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/subset_vars.py
new file mode 100644
index 0000000..6003513
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/scanvi/subset_vars.py
@@ -0,0 +1,31 @@
+def subset_vars(adata, subset_col):
+    """Subset AnnData object on highly variable genes
+
+    Parameters
+    ----------
+    adata : AnnData
+        Annotated data object
+    subset_col : str
+        Name of the boolean column in `adata.var` that contains the information if features should be used or not
+
+    Returns
+    -------
+    AnnData
+        Copy of `adata` with subsetted features
+    """
+    if subset_col not in adata.var.columns:
+        raise ValueError(
+            f"Requested to use .var column '{subset_col}' as a selection of genes, but the column is not available."
+        )
+
+    if adata.var[subset_col].dtype == "boolean":
+        assert adata.var[subset_col].isna().sum() == 0, (
+            f"The .var column `{subset_col}` contains NaN values. Can not subset data."
+        )
+        adata.var[subset_col] = adata.var[subset_col].astype("bool")
+
+    assert adata.var[subset_col].dtype == "bool", (
+        f"Expected dtype of .var column '{subset_col}' to be `bool`, but found {adata.var[subset_col].dtype}. Can not subset data."
+    )
+
+    return adata[:, adata.var[subset_col]].copy()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/.config.vsh.yaml
new file mode 100644
index 0000000..64e82a0
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/.config.vsh.yaml
@@ -0,0 +1,306 @@
+name: "leiden"
+namespace: "cluster"
+version: "v4.0.4"
+authors:
+- name: "Dries De Maeyer"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input file."
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsp_connectivities"
+    description: "In which .obsp slot the neighbor connectivities can be found."
+    info: null
+    default:
+    - "connectivities"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_name"
+    description: "Name of the .obsm key under which to add the cluster labels.\nThe\
+      \ name of the columns in the matrix will correspond to the resolutions.\n"
+    info: null
+    default:
+    - "leiden"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--resolution"
+    description: "A parameter value controlling the coarseness of the clustering.\
+      \ Higher values lead to more clusters.\nMultiple values will result in clustering\
+      \ being performed multiple times.\n"
+    info: null
+    default:
+    - 1.0
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Cluster cells using the [Leiden algorithm] [Traag18] implemented in\
+  \ the [Scanpy framework] [Wolf18]. \nLeiden is an improved version of the [Louvain\
+  \ algorithm] [Blondel08]. \nIt has been proposed for single-cell analysis by [Levine15]\
+  \ [Levine15]. \nThis requires having ran `neighbors/find_neighbors` or `neighbors/bbknn`\
+  \ first.\n\n[Blondel08]: Blondel et al. (2008), Fast unfolding of communities in\
+  \ large networks, J. Stat. Mech.  \n[Levine15]: Levine et al. (2015), Data-Driven\
+  \ Phenotypic Dissection of AML Reveals Progenitor-like Cells that Correlate with\
+  \ Prognosis, Cell.  \n[Traag18]: Traag et al. (2018), From Louvain to Leiden: guaranteeing\
+  \ well-connected communities arXiv.  \n[Wolf18]: Wolf et al. (2018), Scanpy: large-scale\
+  \ single-cell gene expression data analysis, Genome Biology.  \n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "highcpu"
+    - "midmem"
+    - "middisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.13-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    - "leidenalg~=0.10.0"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/cluster/leiden/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/cluster/leiden"
+  executable: "target/nextflow/cluster/leiden/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/main.nf
new file mode 100644
index 0000000..36b2a44
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/main.nf
@@ -0,0 +1,4274 @@
+// leiden v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries De Maeyer (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "leiden",
+  "namespace" : "cluster",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input file.",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsp_connectivities",
+          "description" : "In which .obsp slot the neighbor connectivities can be found.",
+          "default" : [
+            "connectivities"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_name",
+          "description" : "Name of the .obsm key under which to add the cluster labels.\nThe name of the columns in the matrix will correspond to the resolutions.\n",
+          "default" : [
+            "leiden"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--resolution",
+          "description" : "A parameter value controlling the coarseness of the clustering. Higher values lead to more clusters.\nMultiple values will result in clustering being performed multiple times.\n",
+          "default" : [
+            1.0
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Cluster cells using the [Leiden algorithm] [Traag18] implemented in the [Scanpy framework] [Wolf18]. \nLeiden is an improved version of the [Louvain algorithm] [Blondel08]. \nIt has been proposed for single-cell analysis by [Levine15] [Levine15]. \nThis requires having ran `neighbors/find_neighbors` or `neighbors/bbknn` first.\n\n[Blondel08]: Blondel et al. (2008), Fast unfolding of communities in large networks, J. Stat. Mech.  \n[Levine15]: Levine et al. (2015), Data-Driven Phenotypic Dissection of AML Reveals Progenitor-like Cells that Correlate with Prognosis, Cell.  \n[Traag18]: Traag et al. (2018), From Louvain to Leiden: guaranteeing well-connected communities arXiv.  \n[Wolf18]: Wolf et al. (2018), Scanpy: large-scale single-cell gene expression data analysis, Genome Biology.  \n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "highcpu",
+          "midmem",
+          "middisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.13-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4",
+            "leidenalg~=0.10.0"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/cluster/leiden/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/cluster/leiden",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+import signal
+import os
+import time
+import logging
+import logging.handlers
+import warnings
+import h5py
+import mudata as mu
+import pandas as pd
+import scanpy as sc
+import numpy as np
+import numpy.typing as npt
+import anndata as ad
+from multiprocessing import managers, shared_memory, get_context
+from concurrent.futures import ProcessPoolExecutor, process, as_completed
+from scipy.sparse import csr_matrix
+from pathlib import Path
+import shutil
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsp_connectivities': $( if [ ! -z ${VIASH_PAR_OBSP_CONNECTIVITIES+x} ]; then echo "r'${VIASH_PAR_OBSP_CONNECTIVITIES//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_name': $( if [ ! -z ${VIASH_PAR_OBSM_NAME+x} ]; then echo "r'${VIASH_PAR_OBSM_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resolution': $( if [ ! -z ${VIASH_PAR_RESOLUTION+x} ]; then echo "list(map(float, r'${VIASH_PAR_RESOLUTION//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+
+from compress_h5mu import compress_h5mu
+
+_shared_logger_name = "leiden"
+
+
+# Function to check available space in /dev/shm
+def get_available_shared_memory():
+    shm_path = "/dev/shm"
+    shm_stats = os.statvfs(shm_path)
+    return shm_stats.f_bsize * shm_stats.f_bavail
+
+
+class SharedNumpyMatrix:
+    def __init__(
+        self,
+        shared_memory: shared_memory.SharedMemory,
+        dtype: npt.DTypeLike,
+        shape: tuple[int, int],
+    ) -> None:
+        self._memory = shared_memory
+        self._dtype = dtype
+        self._shape = shape
+
+    @classmethod
+    def from_numpy(
+        cls, memory_manager: managers.SharedMemoryManager, array: npt.ArrayLike
+    ):
+        available_shared_memory = get_available_shared_memory()
+        n_bytes_required = array.nbytes
+        if available_shared_memory < n_bytes_required:
+            raise ValueError(
+                "Not enough shared memory (/dev/shm) is available to load the data. "
+                f"Required amount: {n_bytes_required}, available: {available_shared_memory}."
+            )
+        shm = memory_manager.SharedMemory(size=array.nbytes)
+        array_in_shared_memory = np.ndarray(
+            array.shape, dtype=array.dtype, buffer=shm.buf
+        )
+        # Copy the data into shared memory
+        array_in_shared_memory[:] = array[:]
+        return cls(shm, array.dtype, array.shape)
+
+    def to_numpy(self):
+        return np.ndarray(self._shape, dtype=self._dtype, buffer=self._memory.buf)
+
+    def close(self):
+        self._memory.close()
+
+
+class SharedCsrMatrix:
+    def __init__(
+        self,
+        data: SharedNumpyMatrix,
+        indices: SharedNumpyMatrix,
+        indptr: SharedNumpyMatrix,
+        shape: npt.DTypeLike,
+    ):
+        self._data = data
+        self._indices = indices
+        self._indptr = indptr
+        self._shape = shape
+
+    @classmethod
+    def from_csr_matrix(
+        cls, memory_manager: managers.SharedMemoryManager, csr_matrix_obj: csr_matrix
+    ):
+        return cls(
+            SharedNumpyMatrix.from_numpy(memory_manager, csr_matrix_obj.data),
+            SharedNumpyMatrix.from_numpy(memory_manager, csr_matrix_obj.indices),
+            SharedNumpyMatrix.from_numpy(memory_manager, csr_matrix_obj.indptr),
+            csr_matrix_obj.shape,
+        )
+
+    def to_csr_matrix(self):
+        return csr_matrix(
+            (self._data.to_numpy(), self._indices.to_numpy(), self._indptr.to_numpy()),
+            shape=self._shape,
+            copy=False,
+        )
+
+    def close(self):
+        self._data.close()
+        self._indices.close()
+        self._indptr.close()
+
+
+def create_empty_anndata_with_connectivities(connectivities, obs_names):
+    empty_anndata = ad.AnnData(
+        np.zeros((connectivities.shape[0], 1)), obs=pd.DataFrame(index=list(obs_names))
+    )
+    empty_anndata.obsp["connectivities"] = connectivities
+    return empty_anndata
+
+
+def run_single_resolution(shared_csr_matrix, obs_names, resolution):
+    logger = logging.getLogger(_shared_logger_name)
+    logger.info(
+        "Process with PID '%s' for resolution '%s' started", os.getpid(), resolution
+    )
+    try:
+        connectivities = shared_csr_matrix.to_csr_matrix()
+        adata = create_empty_anndata_with_connectivities(connectivities, obs_names)
+        with warnings.catch_warnings():
+            # In the future, the default backend for leiden will be igraph instead of leidenalg.
+            warnings.simplefilter(action="ignore", category=FutureWarning)
+            adata_out = sc.tl.leiden(
+                adata,
+                resolution=resolution,
+                key_added=str(resolution),
+                obsp="connectivities",
+                copy=True,
+            )
+        logger.info(f"Returning result for resolution {resolution}")
+        return adata_out.obs[str(resolution)]
+    finally:
+        obs_names.shm.close()
+        shared_csr_matrix.close()
+
+
+def init_worker(parent_process_id, exit_event, log_queue, log_level):
+    import os
+    import threading
+    import time
+
+    pid = os.getpid()
+
+    logger = logging.getLogger(_shared_logger_name)
+    logger.setLevel(log_level)
+
+    handler = logging.handlers.QueueHandler(log_queue)
+    logger.addHandler(handler)
+
+    logger.info("Initializing process %s", pid)
+
+    def exit_if_orphaned():
+        logger.info(
+            "Starting orphanned process checker for process %s, parent process %s.",
+            pid,
+            parent_process_id,
+        )
+        while True:
+            # Check if parent process is gone
+            try:
+                # If sig is 0, then no signal is sent, but error checking is still performed;
+                # this can be used to check for the existence of a process ID
+                os.kill(parent_process_id, 0)
+            except ProcessLookupError:
+                logger.info("Parent process is gone, shutting down %s", pid)
+                # Kill self
+                os.kill(pid, signal.SIGTERM)
+            time.sleep(0.2)
+            # Parent process requested exit
+            try:
+                exit_event_set = exit_event.wait(timeout=1)
+            except BrokenPipeError:
+                logger.info(
+                    "Checking for shutdown resulted in BrokenPipeError, "
+                    "parent process is most likely gone. Shutting down %s",
+                    pid,
+                )
+                os.kill(pid, signal.SIGTERM)
+            else:
+                if exit_event_set:
+                    logger.info("Exit event set, shutting down %s", pid)
+                    os.kill(pid, signal.SIGTERM)
+            time.sleep(1)
+
+    threading.Thread(target=exit_if_orphaned, daemon=True).start()
+    logger.info(
+        "Initialization of process %s is complete, process is now waiting for work.",
+        pid,
+    )
+
+
+def main():
+    with managers.SyncManager() as syncm:
+        log_level = logging.INFO
+        log_format = "%(name)s:%(levelname)s:%(asctime)s: %(message)s"
+        formatter = logging.Formatter(log_format)
+        console_handler = logging.StreamHandler(sys.stdout)
+        console_handler.setFormatter(formatter)
+        log_queue = syncm.Queue()
+        log_listener = logging.handlers.QueueListener(log_queue, console_handler)
+        log_listener.start()
+
+        logger = logging.getLogger(_shared_logger_name)
+        logger.setLevel(log_level)
+        handler = logging.handlers.QueueHandler(log_queue)
+        logger.addHandler(handler)
+
+        logger.info("Reading %s.", par["input"])
+        adata = mu.read_h5ad(par["input"], mod=par["modality"], backed="r")
+        logger.info("Processing modality '%s'.", par["modality"])
+        try:
+            connectivities = adata.obsp[par["obsp_connectivities"]]
+        except KeyError:
+            raise ValueError(
+                f'Could not find .obsp key "{par["obsp_connectivities"]}" '
+                "in modality {par['modality']}"
+            )
+
+        # An event that, when triggered, will kill the child processes that are still running
+        exit_early_event = syncm.Event()
+        with managers.SharedMemoryManager() as smm:
+            # anndata converts the index to strings, so no worries that it cannot be stored in ShareableList
+            # because it has an unsupported dtype. It should always be string...
+            index_contents = adata.obs.index.to_list()
+            assert all([isinstance(item, str) for item in index_contents])
+            obs_names = smm.ShareableList(index_contents)
+
+            shared_csr_matrix = SharedCsrMatrix.from_csr_matrix(smm, connectivities)
+            results = {}
+            n_workers = (
+                meta["cpus"] - 2 if (meta["cpus"] and (meta["cpus"] - 2) > 0) else 1
+            )
+            logger.info(f"Requesting {n_workers} workers")
+            executor = ProcessPoolExecutor(
+                max_workers=n_workers,
+                max_tasks_per_child=1,
+                mp_context=get_context("spawn"),
+                initializer=init_worker,
+                initargs=((os.getpid(), exit_early_event, log_queue, log_level)),
+            )
+            pending_futures = {
+                executor.submit(
+                    run_single_resolution, shared_csr_matrix, obs_names, resolution
+                ): resolution
+                for resolution in par["resolution"]
+            }
+            try:
+                logger.info("All futures sheduled")
+                for done_future in as_completed(pending_futures):
+                    resolution = pending_futures[done_future]
+                    data = done_future.result()
+                    logger.info(f"Processed resolution '{resolution}'")
+                    results[str(resolution)] = data
+            except process.BrokenProcessPool:
+                # This assumes that one of the child processses was killed by the kernel
+                # because the oom killer was activated. This the is the most likely scenario,
+                # other causes could be:
+                # * Subprocess terminates without raising a proper exception.
+                # * The code of the process handling the communication is broke (i.e. a python bug)
+                # * The return data could not be pickled.
+                logger.error("BrokenProcessPool is raised")
+                executor.shutdown(wait=False, cancel_futures=True)
+                time.sleep(3)
+                exit_early_event.set()
+                time.sleep(3)
+                sys.exit(137)
+            finally:
+                logger.info("Closing shared resources in main process")
+                shared_csr_matrix.close()
+                obs_names.shm.close()
+                logger.info("Shared resources closed")
+                logger.info(
+                    "Shutting down logging queue and re-enabling logging from main thread."
+                )
+                log_listener.enqueue_sentinel()
+                log_listener.stop()
+                stdout_handler = logging.StreamHandler()
+                stdout_handler.setFormatter(formatter)
+                logger.addHandler(stdout_handler)
+                logger.info("Re-enabled logging in main thread.")
+            logger.info("Waiting for shutdown of processes")
+            executor.shutdown()
+            logger.info("Executor shut down.")
+        del adata
+        results = pd.DataFrame(results)
+
+        output_file = Path(par["output"])
+        logger.info("Writing output to %s.", par["output"])
+        output_file_uncompressed = (
+            output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+            if par["output_compression"]
+            else output_file
+        )
+        shutil.copyfile(par["input"], output_file_uncompressed)
+        logger.info("Opening %s", output_file_uncompressed)
+        with h5py.File(output_file_uncompressed, "a") as storage:
+            group_path = f"/mod/{par['modality']}/obsm/{par['obsm_name']}"
+            logger.info("Adding output to %s", group_path)
+            ad.io.write_elem(storage, k=group_path, elem=results)
+        if par["output_compression"]:
+            compress_h5mu(
+                output_file_uncompressed,
+                output_file,
+                compression=par["output_compression"],
+            )
+            output_file_uncompressed.unlink()
+        logger.info("Finished.")
+        log_listener.enqueue_sentinel()
+        time.sleep(3)
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/cluster/leiden",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "highcpu",
+    "midmem",
+    "middisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/nextflow.config
new file mode 100644
index 0000000..21db855
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'cluster/leiden'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Cluster cells using the [Leiden algorithm] [Traag18] implemented in the [Scanpy framework] [Wolf18]. \nLeiden is an improved version of the [Louvain algorithm] [Blondel08]. \nIt has been proposed for single-cell analysis by [Levine15] [Levine15]. \nThis requires having ran `neighbors/find_neighbors` or `neighbors/bbknn` first.\n\n[Blondel08]: Blondel et al. (2008), Fast unfolding of communities in large networks, J. Stat. Mech.  \n[Levine15]: Levine et al. (2015), Data-Driven Phenotypic Dissection of AML Reveals Progenitor-like Cells that Correlate with Prognosis, Cell.  \n[Traag18]: Traag et al. (2018), From Louvain to Leiden: guaranteeing well-connected communities arXiv.  \n[Wolf18]: Wolf et al. (2018), Scanpy: large-scale single-cell gene expression data analysis, Genome Biology.  \n'
+  author = 'Dries De Maeyer'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/nextflow_schema.json
new file mode 100644
index 0000000..cb2d024
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/nextflow_schema.json
@@ -0,0 +1,87 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "leiden",
+  "description": "Cluster cells using the [Leiden algorithm] [Traag18] implemented in the [Scanpy framework] [Wolf18]. \nLeiden is an improved version of the [Louvain algorithm] [Blondel08]. \nIt has been proposed for single-cell analysis by [Levine15] [Levine15]. \nThis requires having ran `neighbors/find_neighbors` or `neighbors/bbknn` first.\n\n[Blondel08]: Blondel et al. (2008), Fast unfolding of communities in large networks, J. Stat. Mech.  \n[Levine15]: Levine et al. (2015), Data-Driven Phenotypic Dissection of AML Reveals Progenitor-like Cells that Correlate with Prognosis, Cell.  \n[Traag18]: Traag et al. (2018), From Louvain to Leiden: guaranteeing well-connected communities arXiv.  \n[Wolf18]: Wolf et al. (2018), Scanpy: large-scale single-cell gene expression data analysis, Genome Biology.  \n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input file.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "obsp_connectivities": {
+          "type": "string",
+          "description": "In which .obsp slot the neighbor connectivities can be found.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"connectivities\"`. ",
+          "default": "connectivities"
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "obsm_name": {
+          "type": "string",
+          "description": "Name of the .obsm key under which to add the cluster labels.\nThe name of the columns in the matrix will correspond to the resolutions.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"leiden\"`. ",
+          "default": "leiden"
+        },
+        "resolution": {
+          "type": "array",
+          "items": {
+            "type": "number"
+          },
+          "description": "A parameter value controlling the coarseness of the clustering",
+          "help_text": "Type: `double`, multiple: `True`, required, default: `[1.0]`. ",
+          "default": [
+            1.0
+          ]
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/cluster/leiden/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/.config.vsh.yaml
new file mode 100644
index 0000000..5a00be4
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/.config.vsh.yaml
@@ -0,0 +1,293 @@
+name: "from_cellranger_multi_to_h5mu"
+namespace: "convert"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input folder. Must contain the output from a cellranger multi run."
+    info: null
+    example:
+    - "input_dir_containing_modalities"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Locations for the output files. Must contain a wildcard (*) character,\n\
+      which will be replaced with the sample name.\n"
+    info: null
+    example:
+    - "*.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--sample_csv"
+    description: "CSV file describing the sample name per output file"
+    info: null
+    example:
+    - "samples.csv"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--uns_metrics"
+    description: "Name of the .uns slot under which to QC metrics (if any)."
+    info: null
+    default:
+    - "metrics_cellranger"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Converts the output from cellranger multi to a single .h5mu file.\n\
+  By default, will map the following library type names to modality names:\n  - Gene\
+  \ Expression: rna\n  - Peaks: atac\n  - Antibody Capture: prot\n  - VDJ: vdj\n \
+  \ - VDJ-T: vdj_t\n  - VDJ-B: vdj_b\n  - CRISPR Guide Capture: crispr\n  - Multiplexing\
+  \ Capture: hashing\n  \nOther library types have their whitepace removed and dashes\
+  \ replaced by\nunderscores to generate the modality name.\n\nCurrently does not\
+  \ allow parsing the output from cell barcode demultiplexing.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "10x_5k_anticmv"
+- type: "file"
+  path: "10x_5k_anticmv_v10"
+- type: "file"
+  path: "10x_5k_lung_crispr"
+- type: "file"
+  path: "10x_5k_lung_crispr_v10"
+- type: "file"
+  path: "10x_5k_beam"
+- type: "file"
+  path: "10x_5k_beam_v10"
+- type: "file"
+  path: "10x_5k_fixed"
+- type: "file"
+  path: "10x_5k_fixed_v10"
+- type: "file"
+  path: "10x_4plex_dtc"
+- type: "file"
+  path: "10x_4plex_dtc_v10"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "lowmem"
+    - "singlecpu"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.13-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    - "scirpy"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/convert/from_cellranger_multi_to_h5mu/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/convert/from_cellranger_multi_to_h5mu"
+  executable: "target/nextflow/convert/from_cellranger_multi_to_h5mu/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/main.nf
new file mode 100644
index 0000000..313d029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/main.nf
@@ -0,0 +1,4408 @@
+// from_cellranger_multi_to_h5mu v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "from_cellranger_multi_to_h5mu",
+  "namespace" : "convert",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input folder. Must contain the output from a cellranger multi run.",
+          "example" : [
+            "input_dir_containing_modalities"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Locations for the output files. Must contain a wildcard (*) character,\nwhich will be replaced with the sample name.\n",
+          "example" : [
+            "*.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--sample_csv",
+          "description" : "CSV file describing the sample name per output file",
+          "example" : [
+            "samples.csv"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--uns_metrics",
+          "description" : "Name of the .uns slot under which to QC metrics (if any).",
+          "default" : [
+            "metrics_cellranger"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Converts the output from cellranger multi to a single .h5mu file.\nBy default, will map the following library type names to modality names:\n  - Gene Expression: rna\n  - Peaks: atac\n  - Antibody Capture: prot\n  - VDJ: vdj\n  - VDJ-T: vdj_t\n  - VDJ-B: vdj_b\n  - CRISPR Guide Capture: crispr\n  - Multiplexing Capture: hashing\n  \nOther library types have their whitepace removed and dashes replaced by\nunderscores to generate the modality name.\n\nCurrently does not allow parsing the output from cell barcode demultiplexing.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_anticmv"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_anticmv_v10"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_lung_crispr"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_lung_crispr_v10"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_beam"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_beam_v10"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_fixed"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_fixed_v10"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_4plex_dtc"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_4plex_dtc_v10"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "lowmem",
+          "singlecpu"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.13-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4",
+            "scirpy"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/convert/from_cellranger_multi_to_h5mu/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/convert/from_cellranger_multi_to_h5mu",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+from pathlib import Path
+import sys
+import scanpy
+import pandas as pd
+import mudata
+import anndata
+import numpy as np
+from scirpy.io import read_10x_vdj
+from collections import defaultdict
+from functools import partial
+import json
+import csv
+import tempfile
+
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'sample_csv': $( if [ ! -z ${VIASH_PAR_SAMPLE_CSV+x} ]; then echo "r'${VIASH_PAR_SAMPLE_CSV//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'uns_metrics': $( if [ ! -z ${VIASH_PAR_UNS_METRICS+x} ]; then echo "r'${VIASH_PAR_UNS_METRICS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+POSSIBLE_LIBRARY_TYPES = (
+    "vdj_t",
+    "vdj_b",
+    "vdj_t_gd",
+    "count",
+    "antigen_analysis",
+    "multiplexing_analysis",
+)
+
+FEATURE_TYPES_NAMES = {
+    "Gene Expression": "rna",
+    "Peaks": "atac",
+    "Antibody Capture": "prot",
+    "VDJ": "vdj",
+    "VDJ-T": "vdj_t",
+    "VDJ-B": "vdj_b",
+    "CRISPR Guide Capture": "gdo",
+    "Multiplexing Capture": "hto",
+    "Antigen Capture": "antigen",
+    "Custom": "custom",
+}
+
+
+def cast_to_writeable_dtype(result: pd.DataFrame) -> pd.DataFrame:
+    """
+    Cast the dataframe to dtypes that can be written by mudata.
+    """
+    # dtype inferral workfs better with np.nan
+    result = result.replace({pd.NA: np.nan})
+
+    # MuData supports nullable booleans and ints
+    # ie. \\`IntegerArray\\` and \\`BooleanArray\\`
+    result = result.convert_dtypes(
+        infer_objects=True,
+        convert_integer=True,
+        convert_string=False,
+        convert_boolean=True,
+        convert_floating=False,
+    )
+
+    # Convert leftover 'object' columns to string
+    # However, na values are supported, so convert all values except NA's to string
+    object_cols = result.select_dtypes(include="object").columns.values
+    for obj_col in object_cols:
+        result[obj_col] = (
+            result[obj_col]
+            .where(result[obj_col].isna(), result[obj_col].astype(str))
+            .astype("category")
+        )
+    return result
+
+
+def gather_input_data(dir: Path):
+    # /
+    # +-- multi
+    # |    +-- count (raw output)
+    # |    |    +-- feature_reference.csv
+    # |    |    +-- raw_feature_bc_matrix.h5
+    # |    +-- vdj_t
+    # |    |    +-- all_contig_annotations.json
+    # |    +-- vdj_b
+    # |    |    +-- all_contig_annotations.json
+    # |    +-- vdj_t_gd
+    # |    |    +-- all_contig_annotations.json
+    # |    +--  multiplexing_analysis
+    # |         +-- cells_per_tag.json
+    # +-- per_sample_outs (filtered outputs)
+    #      +-- example_1
+    #           +-- antigen_analysis
+    #           |   +-- per_barcode.csv
+    #           |   +-- antigen_specificity_scores.csv
+    #           +-- count
+    #           |   +-- antibody_analysis
+    #           |   +-- crispr_analysis
+    #           |       +-- perturbation_efficiencies_by_feature.csv
+    #           |       +-- perturbation_efficiencies_by_target.csv
+    #           +-- vdj_t (unused)
+    #           +-- vdj_b (unused)
+    #           +-- vdj_t_gd (unused)
+    #           +-- metrics_summary.csv
+
+    if not dir.is_dir():
+        raise ValueError("Specified input is not a directory.")
+    folder_contents = list(dir.iterdir())
+    config = dir / "config.csv"
+    if config not in folder_contents:
+        logger.warning(
+            "Config.csv not found in input directory, this folder might not be a valid cellranger multi output."
+        )
+
+    required_subfolders = [
+        dir / subfolder_name for subfolder_name in ("per_sample_outs",)
+    ]
+    found_input = {key_: {} for key_ in POSSIBLE_LIBRARY_TYPES}
+    for required_subfolder in required_subfolders:
+        if required_subfolder not in folder_contents:
+            raise ValueError(
+                f"Input folder must contain the subfolder {required_subfolder} please make "
+                "sure that the specified input folder is a valid cellranger multi output."
+            )
+
+    library_type_dir = multi_dir if (multi_dir := (dir / "multi")).is_dir() else dir
+    for library_type in library_type_dir.iterdir():
+        if library_type.name in POSSIBLE_LIBRARY_TYPES:
+            if multi_dir.is_dir():
+                if not library_type.is_dir():
+                    logger.warning(
+                        "%s is not a directory. Contents of the multi folder "
+                        "must be directories to be recognized as valid input data",
+                        library_type,
+                    )
+                    continue
+            found_input[library_type.name] = library_type
+            continue
+        if multi_dir.is_dir():
+            raise ValueError(
+                f"Contents of the 'multi' folder must be found one of the following: {','.join(POSSIBLE_LIBRARY_TYPES)}."
+            )
+    if not found_input["count"]:
+        found_input["count"] = dir
+
+    feature_reference = found_input["count"] / "feature_reference.csv"
+    if feature_reference.is_file():
+        found_input["feature_reference"] = feature_reference
+
+    per_sample_outs_dir = dir / "per_sample_outs"
+    samples_dirs = [
+        samplepath
+        for samplepath in per_sample_outs_dir.iterdir()
+        if samplepath.is_dir()
+    ]
+    for samples_dir in samples_dirs:
+        for file_part in (
+            "metrics_summary.csv",
+            "count/crispr_analysis/perturbation_efficiencies_by_feature.csv",
+            "crispr_analysis/perturbation_efficiencies_by_feature.csv",  # Cell Ranger v10
+            "count/crispr_analysis/perturbation_efficiencies_by_target.csv",
+            "crispr_analysis/perturbation_efficiencies_by_target.csv",  # Cell Ranger v10
+            "antigen_analysis",
+        ):
+            found_file = samples_dir / file_part
+            if found_file.exists():
+                file_name = found_file.name.removesuffix(".csv")
+                found_input.setdefault(file_name, {})[samples_dir.name] = found_file
+
+    return found_input
+
+
+def proces_perturbation(
+    key_name: str, mudatas: dict[str, mudata.MuData], efficiency_files: dict[str, Path]
+):
+    for sample_name, mudata_obj in mudatas.items():
+        efficiency_file = efficiency_files[sample_name]
+        assert "gdo" in mudata_obj.mod
+        eff_df = pd.read_csv(
+            efficiency_file,
+            index_col="Perturbation",
+            sep=",",
+            decimal=".",
+            quotechar='"',
+        )
+        mudata_obj.mod["gdo"].uns[key_name] = eff_df
+    return mudatas
+
+
+def process_feature_reference(
+    mudatas: dict[str, mudata.MuData], efficiency_file: dict[str, Path]
+):
+    df_all = pd.read_csv(
+        efficiency_file, index_col="id", sep=",", decimal=".", quotechar='"'
+    )
+    assert "feature_type" in df_all.columns, (
+        "Columns 'feature_type' should be present in features_reference file."
+    )
+    for _, mudata_obj in mudatas.items():
+        df = df_all.copy(deep=True)
+        feature_types = df["feature_type"]
+        missing_features = set(feature_types) - set(FEATURE_TYPES_NAMES)
+        if missing_features:
+            raise ValueError(
+                "Not all feature types present in the features_reference file are supported by this component.\\\\n"
+                f"Missing support for features: {','.join(missing_features)}."
+            )
+        for feature_type in feature_types:
+            modality = FEATURE_TYPES_NAMES[feature_type]
+            subset_df = df.loc[df["feature_type"] == feature_type]
+            mudata_obj.mod[modality].uns["feature_reference"] = subset_df
+    return mudatas
+
+
+def process_counts(counts_folder: Path, multiplexing_info, metrics_files):
+    counts_matrix_file = counts_folder / "raw_feature_bc_matrix.h5"
+    logger.info("Reading %s.", counts_matrix_file)
+    adata = scanpy.read_10x_h5(counts_matrix_file, gex_only=False)
+
+    # set the gene ids as var_names
+    logger.info("Renaming var columns")
+    adata.var = adata.var.rename_axis("gene_symbol").reset_index().set_index("gene_ids")
+
+    # generate output
+    logger.info("Convert to mudata")
+
+    def modality_name_factory(library_type):
+        return ("".join(library_type.replace("-", "_").split())).lower()
+
+    feature_types = defaultdict(modality_name_factory, FEATURE_TYPES_NAMES)
+    mudata_all_samples = mudata.MuData(adata, feature_types_names=feature_types)
+    if multiplexing_info:
+        # Get the mapping between the barcode and the sample ID from one of the metrics files
+        metrics_file = pd.read_csv(
+            list(metrics_files.values())[0], decimal=".", quotechar='"', thousands=","
+        )
+        sample_ids = metrics_file[metrics_file["Metric Name"] == "Sample ID"]
+        barcode_sample_mapping = (
+            sample_ids.loc[:, ["Group Name", "Metric Value"]]
+            .set_index("Group Name")
+            .squeeze()
+            .to_dict()
+        )
+        # When probe barcodes are combined with multiplexing barcodes; the probe barcode (BC)
+        # is paired with other sample with other sample information using the syntax "{BC}+{antibody}"
+        # (or even "{BC}+{antibody}+{crispr guide}"). The cells_per_tag
+        # file only encodes the BCs; so we need to strip the other information.
+        barcode_sample_mapping = {
+            key_.partition("+")[0]: value_
+            for key_, value_ in barcode_sample_mapping.items()
+        }
+        return split_samples(
+            mudata_all_samples, multiplexing_info, barcode_sample_mapping
+        )
+    return {"run": mudata_all_samples}
+
+
+def split_samples(mudata_obj, multiplexing_analysis_folder, barcode_sample_mapping):
+    result = {}
+    cells_per_tag_file = multiplexing_analysis_folder / "cells_per_tag.json"
+    with cells_per_tag_file.open("r") as open_json:
+        sample_cell_mapping = json.load(open_json)
+
+    for barcode, indices in sample_cell_mapping.items():
+        if indices:
+            sample_mudata = mudata_obj[indices]
+            sample = barcode_sample_mapping[barcode]
+            result[sample] = sample_mudata.copy()
+    return result
+
+
+def process_metrics_summary(
+    mudatas: dict[str, mudata.MuData], metrics_files: dict[str, Path]
+):
+    def read_percentage(val):
+        try:
+            if str(val).endswith("%"):
+                return float(val.strip("%")) / 100
+            else:
+                return val
+        except (AttributeError, ValueError):
+            return val
+
+    for sample, mudata_obj in mudatas.items():
+        metrics_file = metrics_files[sample]
+        metrics_summary = pd.read_csv(
+            metrics_file, decimal=".", quotechar='"', thousands=","
+        ).applymap(read_percentage)
+
+        mudata_obj.uns[par["uns_metrics"]] = metrics_summary
+        for colname, coldata in metrics_summary.items():
+            try:
+                new_column = coldata.astype(str, copy=True).astype(
+                    {colname: "category"}
+                )
+                metrics_summary[colname] = new_column
+            except (ValueError, TypeError):
+                logger.warning(f"Could not store column {colname} from metrics.")
+                pass
+    return mudatas
+
+
+def process_antigen_analysis(
+    mudatas: dict[str, mudata.MuData], antigen_analysis_folder_paths: dict[str, Path]
+):
+    for sample_id, mudata_obj in mudatas.items():
+        antigen_analysis_folder_path = antigen_analysis_folder_paths[sample_id]
+        assert "antigen" in mudata_obj.mod
+        per_barcodes_file = antigen_analysis_folder_path / "per_barcode.csv"
+        assert per_barcodes_file.is_file(), (
+            "Expected a per_barcode.csv file to be present."
+        )
+        per_barcodes_df = pd.read_csv(
+            per_barcodes_file, index_col="barcode", sep=",", decimal=".", quotechar='"'
+        )
+        is_gex_cell = per_barcodes_df["is_gex_cell"]
+        assert len(set(is_gex_cell.unique().tolist()) - set([False, True])) == 0, (
+            "Expected 'is_gex_cell' column to be boolean. Please report this as a bug."
+        )
+        barcodes_in_gex = per_barcodes_df[is_gex_cell]
+        # All of the barcodes listed in the per_barcode.csv with is_gex_cell set to 'True'
+        # must be in the 'rna' (an thus also 'antigen') modality
+        assert barcodes_in_gex.index.difference(mudata_obj["rna"].obs_names).empty
+        orig_obs_names = mudata_obj["antigen"].obs_names.copy()
+        mudata_obj["antigen"].obs = cast_to_writeable_dtype(
+            pd.concat(
+                [mudata_obj["antigen"].obs, barcodes_in_gex],
+                axis="columns",
+                join="outer",
+                verify_integrity=True,
+                sort=False,
+            )
+        )
+        assert orig_obs_names.equals(mudata_obj["antigen"].obs_names)
+        del orig_obs_names
+
+        # The antigen_specificity_scores.csv file is only present when cellranger
+        # multi was run with a [antigen-specificity] section in config
+        specificity_file = (
+            antigen_analysis_folder_path / "antigen_specificity_scores.csv"
+        )
+        if specificity_file.is_file():
+            antigen_scores_df = pd.read_csv(
+                specificity_file,
+                index_col=["barcode", "antigen"],
+                sep=",",
+                decimal=".",
+                quotechar='"',
+            )
+            score = antigen_scores_df.unstack()
+            assert score.index.difference(mudata_obj["rna"].obs_names).empty
+            antigens = score.columns.unique(level="antigen")
+            for antigen in antigens:
+                score_antigen = score.loc[:, (slice(None), antigen)].droplevel(
+                    "antigen", axis=1
+                )
+                score_antigen = score_antigen.reindex(mudata_obj["rna"].obs_names)
+                mudata_obj["antigen"].obsm[f"antigen_specificity_scores_{antigen}"] = (
+                    cast_to_writeable_dtype(score_antigen)
+                )
+    return mudatas
+
+
+def process_vdj(mudatas: dict[str, mudata.MuData], vdj_folder_path: str):
+    # https://scverse.org/scirpy/latest/generated/scirpy.io.read_10x_vdj.html#scirpy-io-read-10x-vdj
+    # According to docs, using the json is preferred as this file includes intron info.
+    all_config_json_file = vdj_folder_path / "all_contig_annotations.json"
+    vdj_type = vdj_folder_path.name
+    with all_config_json_file.open("r") as open_json:
+        json_obj = json.load(open_json)
+    for _, mudata_obj in mudatas.items():
+        vdj_anndata = anndata.AnnData()
+        json_for_sample = [
+            entry for entry in json_obj if entry["barcode"] in mudata_obj.obs_names
+        ]
+        if json_for_sample:
+            with tempfile.NamedTemporaryFile(mode="w", suffix=".json") as tfile:
+                json.dump(json_for_sample, tfile, indent=4)
+                tfile.flush()
+                vdj_anndata = read_10x_vdj(tfile.name)
+        mudata_obj.mod[vdj_type] = vdj_anndata
+    return mudatas
+
+
+def get_modalities(input_data):
+    dispatcher = {
+        "multiplexing_analysis": split_samples,
+        "vdj_t": process_vdj,
+        "vdj_b": process_vdj,
+        "vdj_t_gd": process_vdj,
+        "metrics_summary": process_metrics_summary,
+        "feature_reference": process_feature_reference,
+        "perturbation_efficiencies_by_feature": partial(
+            proces_perturbation, "perturbation_efficiencies_by_feature"
+        ),
+        "perturbation_efficiencies_by_target": partial(
+            proces_perturbation, "perturbation_efficiencies_by_target"
+        ),
+        "antigen_analysis": process_antigen_analysis,
+    }
+    mudata_per_sample = process_counts(
+        input_data["count"],
+        input_data["multiplexing_analysis"],
+        input_data["metrics_summary"],
+    )
+    for modality_name, modality_data_path in input_data.items():
+        if (
+            modality_name in ("count", "multiplexing_analysis")
+            or not modality_data_path
+        ):
+            continue
+        try:
+            parser_function = dispatcher[modality_name]
+        except KeyError as e:
+            raise ValueError(
+                "This component does not support the "
+                f"parsing of the '{modality_name}' yet."
+            ) from e
+        mudata_per_sample = parser_function(mudata_per_sample, modality_data_path)
+    return mudata_per_sample
+
+
+def main():
+    cellranger_multi_dir = Path(par["input"])
+    # TODO: remove when issue https://github.com/viash-io/viash/issues/706 is resolved.
+    if isinstance(par["output"], (list, set, tuple)):
+        assert len(par["output"]) == 1, (
+            "A single output file template should have been provided."
+        )
+        par["output"] = par["output"][0]
+    assert par["output"].count("*") == 1, (
+        f"Expected exactly one wildcard character (*) in output "
+        f"files template ({par['output']}). Found {par['output'].count('*')}"
+    )
+    input_data = gather_input_data(cellranger_multi_dir)
+    result = get_modalities(input_data)
+    output_files = {
+        par["output"].replace("*", sample_name) for sample_name in result.keys()
+    }
+    assert len(output_files) == len(result.keys()), (
+        "Replacing the wildcard in the output files "
+        "template did not produce unique file paths."
+    )
+    logger.info(
+        "Writing output for samples: '%s' to '%s'",
+        "".join(result.keys()),
+        par["output"],
+    )
+    with Path(par["sample_csv"]).open("w", newline="") as open_csv:
+        csvwriter = csv.DictWriter(open_csv, fieldnames=["sample_name", "file"])
+        csvwriter.writeheader()
+        for sample_name, mudata_obj in result.items():
+            output_file = Path(par["output"].replace("*", sample_name))
+            mudata_obj.write_h5mu(output_file, compression=par["output_compression"])
+            csvwriter.writerow({"sample_name": sample_name, "file": output_file.name})
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/convert/from_cellranger_multi_to_h5mu",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "lowmem",
+    "singlecpu"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/nextflow.config
new file mode 100644
index 0000000..6ed99a1
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'convert/from_cellranger_multi_to_h5mu'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Converts the output from cellranger multi to a single .h5mu file.\nBy default, will map the following library type names to modality names:\n  - Gene Expression: rna\n  - Peaks: atac\n  - Antibody Capture: prot\n  - VDJ: vdj\n  - VDJ-T: vdj_t\n  - VDJ-B: vdj_b\n  - CRISPR Guide Capture: crispr\n  - Multiplexing Capture: hashing\n  \nOther library types have their whitepace removed and dashes replaced by\nunderscores to generate the modality name.\n\nCurrently does not allow parsing the output from cell barcode demultiplexing.\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/nextflow_schema.json
new file mode 100644
index 0000000..dcdb421
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/nextflow_schema.json
@@ -0,0 +1,74 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "from_cellranger_multi_to_h5mu",
+  "description": "Converts the output from cellranger multi to a single .h5mu file.\nBy default, will map the following library type names to modality names:\n  - Gene Expression: rna\n  - Peaks: atac\n  - Antibody Capture: prot\n  - VDJ: vdj\n  - VDJ-T: vdj_t\n  - VDJ-B: vdj_b\n  - CRISPR Guide Capture: crispr\n  - Multiplexing Capture: hashing\n  \nOther library types have their whitepace removed and dashes replaced by\nunderscores to generate the modality name.\n\nCurrently does not allow parsing the output from cell barcode demultiplexing.\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input folder",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input_dir_containing_modalities\"`. "
+        },
+        "output": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "Locations for the output files",
+          "help_text": "Type: `file`, multiple: `True`, default: `\"$id.$key.output_*.h5mu\"`, direction: `output`, example: `[\"*.h5mu\"]`. ",
+          "default": "$id.$key.output_*.h5mu"
+        },
+        "sample_csv": {
+          "type": "string",
+          "format": "path",
+          "description": "CSV file describing the sample name per output file",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.sample_csv.csv\"`, direction: `output`, example: `\"samples.csv\"`. ",
+          "default": "$id.$key.sample_csv.csv"
+        },
+        "uns_metrics": {
+          "type": "string",
+          "description": "Name of the .uns slot under which to QC metrics (if any).",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"metrics_cellranger\"`. ",
+          "default": "metrics_cellranger"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_cellranger_multi_to_h5mu/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/.config.vsh.yaml
new file mode 100644
index 0000000..e83be69
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/.config.vsh.yaml
@@ -0,0 +1,306 @@
+name: "from_tiledb_to_h5mu"
+namespace: "convert"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Input database"
+  description: "Open a tileDB-SOMA database by URI."
+  arguments:
+  - type: "string"
+    name: "--input_uri"
+    description: "A URI pointing to a TileDB-SOMA database."
+    info: null
+    example:
+    - "s3://bucket/path"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--s3_region"
+    description: "Region where the TileDB-SOMA database is hosted.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--endpoint"
+    description: "Custom endpoint to use to connect to S3\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--s3_no_sign_request"
+    description: "Do not sign S3 requests. Credentials will not be loaded if this\
+      \ argument is provided.\n"
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_modality"
+    description: "TileDB-SOMA measurement to take the input from.\n"
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_layers"
+    description: "Layers to import from TileDB-SOMA database.\n"
+    info: null
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "MuData output"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "The output MuData file.\n"
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_modality"
+    description: "Name of the modality to write the results to.\n"
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "tiledb"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+  docker_run_args:
+  - "--env"
+  - "AWS_ACCESS_KEY_ID"
+  - "--env"
+  - "AWS_SECRET_ACCESS_KEY"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "highmem"
+    - "midcpu"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "tiledbsoma"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  - type: "python"
+    user: false
+    packages:
+    - "moto[server]"
+    - "boto3"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/convert/from_tiledb_to_h5mu/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/convert/from_tiledb_to_h5mu"
+  executable: "target/nextflow/convert/from_tiledb_to_h5mu/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/main.nf
new file mode 100644
index 0000000..a220352
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/main.nf
@@ -0,0 +1,4061 @@
+// from_tiledb_to_h5mu v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author, maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "from_tiledb_to_h5mu",
+  "namespace" : "convert",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Input database",
+      "description" : "Open a tileDB-SOMA database by URI.",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--input_uri",
+          "description" : "A URI pointing to a TileDB-SOMA database.",
+          "example" : [
+            "s3://bucket/path"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--s3_region",
+          "description" : "Region where the TileDB-SOMA database is hosted.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--endpoint",
+          "description" : "Custom endpoint to use to connect to S3\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--s3_no_sign_request",
+          "description" : "Do not sign S3 requests. Credentials will not be loaded if this argument is provided.\n",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_modality",
+          "description" : "TileDB-SOMA measurement to take the input from.\n",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_layers",
+          "description" : "Layers to import from TileDB-SOMA database.\n",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "MuData output",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "The output MuData file.\n",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_modality",
+          "description" : "Name of the modality to write the results to.\n",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/tiledb/"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild",
+      "docker_run_args" : [
+        "--env",
+        "AWS_ACCESS_KEY_ID",
+        "--env",
+        "AWS_SECRET_ACCESS_KEY"
+      ]
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "highmem",
+          "midcpu"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "tiledbsoma"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "moto[server]",
+            "boto3"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/convert/from_tiledb_to_h5mu/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/convert/from_tiledb_to_h5mu",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import tiledbsoma
+import tiledbsoma.io
+import sys
+import mudata
+import json
+import pandas as pd
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input_uri': $( if [ ! -z ${VIASH_PAR_INPUT_URI+x} ]; then echo "r'${VIASH_PAR_INPUT_URI//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  's3_region': $( if [ ! -z ${VIASH_PAR_S3_REGION+x} ]; then echo "r'${VIASH_PAR_S3_REGION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'endpoint': $( if [ ! -z ${VIASH_PAR_ENDPOINT+x} ]; then echo "r'${VIASH_PAR_ENDPOINT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  's3_no_sign_request': $( if [ ! -z ${VIASH_PAR_S3_NO_SIGN_REQUEST+x} ]; then echo "r'${VIASH_PAR_S3_NO_SIGN_REQUEST//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'input_modality': $( if [ ! -z ${VIASH_PAR_INPUT_MODALITY+x} ]; then echo "r'${VIASH_PAR_INPUT_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_layers': $( if [ ! -z ${VIASH_PAR_INPUT_LAYERS+x} ]; then echo "r'${VIASH_PAR_INPUT_LAYERS//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_modality': $( if [ ! -z ${VIASH_PAR_OUTPUT_MODALITY+x} ]; then echo "r'${VIASH_PAR_OUTPUT_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+tiledbsoma.logging.info()
+
+
+def _log_arguments(function_obj, arg_dict):
+    """
+    Format a dictionairy of arguments into a string that is put into the script logs.
+    """
+    args_str = [f"\\\\t{param}: {param_val}\\\\n" for param, param_val in arg_dict.items()]
+    logger.info(
+        "Calling %s with arguments:\\\\n%s",
+        function_obj.__name__,
+        "".join(args_str).rstrip(),
+    )
+
+
+def main(par):
+    logger.info("Component %s started", meta["name"])
+    tiledb_config = {}
+    optional_config = {
+        "vfs.s3.no_sign_request": par["s3_no_sign_request"],
+        "vfs.s3.region": par["s3_region"],
+        "vfs.s3.endpoint_override": par["endpoint"],
+    }
+    for config_setting, config_val in optional_config.items():
+        if config_val is not None:
+            tiledb_config[config_setting] = config_val
+    logger.info("Using the following config to connect: %s", tiledb_config)
+    context = tiledbsoma.SOMATileDBContext(tiledb_config=tiledb_config)
+
+    logger.info(
+        "Trying to access '%s' in region '%s'", par["input_uri"], par["s3_region"]
+    )
+    with tiledbsoma.open(
+        par["input_uri"], mode="r", context=context
+    ) as open_experiment:
+        logger.info("Connection successful")
+        to_anndata_args = {
+            "experiment": open_experiment,
+            "measurement_name": par["input_modality"],
+            "extra_X_layer_names": par["input_layers"],
+        }
+        func_to_call = tiledbsoma.io.to_anndata
+        _log_arguments(func_to_call, to_anndata_args)
+        output_modality = func_to_call(**to_anndata_args)
+        logger.info("Output anndata was:\\\\n%s", output_modality)
+        logger.info("Adding column indices to varm and obsm items.")
+        for multimodal_attribute in ("varm", "obsm"):
+            multimodal_dict = getattr(output_modality, multimodal_attribute)
+            for key_ in list(multimodal_dict.keys()):
+                metadata = getattr(
+                    open_experiment.ms[par["input_modality"]], multimodal_attribute
+                )[key_].metadata
+                if "column_index" in metadata:
+                    logger.info(
+                        "Found index for item %s in %s", key_, multimodal_attribute
+                    )
+                    index_list = json.loads(metadata["column_index"])
+                    assert isinstance(index_list, list)
+                    multimodal_dict[key_] = pd.DataFrame(
+                        multimodal_dict[key_],
+                        columns=index_list,
+                        index=multimodal_dict.dim_names,
+                    )
+                logger.info(
+                    "No column index found for item %s in %s",
+                    key_,
+                    multimodal_attribute,
+                )
+
+    logger.info("Converting to MuData")
+    output_mudata = mudata.MuData({par["output_modality"]: output_modality})
+    logger.info(
+        "Writing output MuData to %s with compression %s",
+        par["output"],
+        par["output_compression"],
+    )
+    output_mudata.write(par["output"], compression=par["output_compression"])
+    logger.info("Finished!")
+
+
+if __name__ == "__main__":
+    main(par)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/convert/from_tiledb_to_h5mu",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "highmem",
+    "midcpu"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/nextflow.config
new file mode 100644
index 0000000..53e5ef2
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/nextflow.config
@@ -0,0 +1,125 @@
+manifest {
+  name = 'convert/from_tiledb_to_h5mu'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/nextflow_schema.json
new file mode 100644
index 0000000..2ea11bb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/nextflow_schema.json
@@ -0,0 +1,100 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "from_tiledb_to_h5mu",
+  "description": "No description",
+  "type": "object",
+  "$defs": {
+    "input database": {
+      "title": "Input database",
+      "type": "object",
+      "description": "Open a tileDB-SOMA database by URI.",
+      "properties": {
+        "input_uri": {
+          "type": "string",
+          "description": "A URI pointing to a TileDB-SOMA database.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"s3://bucket/path\"`. "
+        },
+        "s3_region": {
+          "type": "string",
+          "description": "Region where the TileDB-SOMA database is hosted.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "endpoint": {
+          "type": "string",
+          "description": "Custom endpoint to use to connect to S3\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "s3_no_sign_request": {
+          "type": "boolean",
+          "description": "Do not sign S3 requests",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "input_modality": {
+          "type": "string",
+          "description": "TileDB-SOMA measurement to take the input from.\n",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        },
+        "input_layers": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Layers to import from TileDB-SOMA database.\n",
+          "help_text": "Type: `string`, multiple: `True`, required. "
+        }
+      }
+    },
+    "mudata output": {
+      "title": "MuData output",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "The output MuData file.\n",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "output_modality": {
+          "type": "string",
+          "description": "Name of the modality to write the results to.\n",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/input database"
+    },
+    {
+      "$ref": "#/$defs/mudata output"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/convert/from_tiledb_to_h5mu/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/.config.vsh.yaml
new file mode 100644
index 0000000..1985a3c
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/.config.vsh.yaml
@@ -0,0 +1,348 @@
+name: "concatenate_h5mu"
+namespace: "dataflow"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Paths to the different samples to be concatenated."
+    info: null
+    example:
+    - "sample_paths"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Only output concatenated objects for the provided modalities. Outputs\
+      \ all modalities by default."
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_id"
+    description: "Names of the different samples that have to be concatenated.  Must\
+      \ be specified when using '--mode move'.\nIn this case, the ids will be used\
+      \ for the columns names of the dataframes registring the conflicts.\nIf specified,\
+      \ must be of same length as `--input`.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output location for the concatenated MuData object file.\n"
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_sample_name"
+    description: "Name of the .obs key under which to add the sample names."
+    info: null
+    default:
+    - "sample_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--other_axis_mode"
+    description: "How to handle the merging of other axis (var, obs, ...).\n\n  -\
+      \ None: keep no data\n  - same: only keep elements of the matrices which are\
+      \ the same in each of the samples\n  - unique: only keep elements for which\
+      \ there is only 1 possible value (1 value that can occur in multiple samples)\n\
+      \  - first: keep the annotation from the first sample\n  - only: keep elements\
+      \ that show up in only one of the objects (1 unique element in only 1 sample)\n\
+      \  - move: identical to 'same', but moving the conflicting values to .varm or\
+      \ .obsm\n"
+    info: null
+    default:
+    - "move"
+    required: false
+    choices:
+    - "same"
+    - "unique"
+    - "first"
+    - "only"
+    - "concat"
+    - "move"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--uns_merge_mode"
+    description: "How to handle the merging of .uns across modalities\n  - None: keep\
+      \ no data\n  - same: only keep elements of the matrices which are the same in\
+      \ each of the samples\n  - unique: only keep elements for which there is only\
+      \ 1 possible value (1 value that can occur in multiple samples)\n  - first:\
+      \ keep the annotation from the first sample\n  - only: keep elements that show\
+      \ up in only one of the objects (1 unique element in only 1 sample)\n  - make_unique:\
+      \ identical to 'unique', but keys which are not unique are made unique by prefixing\
+      \ them with the sample id.\n"
+    info: null
+    default:
+    - "make_unique"
+    required: false
+    choices:
+    - "same"
+    - "unique"
+    - "first"
+    - "only"
+    - "make_unique"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsp_keys"
+    description: "List of `.obsp` keys for which block-diagonal concatenation should\
+      \ be performed.\nIf not provided, no `.obsp` keys will be concatenated.\nProvided\
+      \ keys must be present in all samples for block concatenation to be performed.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Concatenate observations from samples in several (uni- and/or multi-modal)\
+  \ MuData files into a single file.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+- type: "file"
+  path: "human_brain_3k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "midcpu"
+    - "highmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.13-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    - "pytest-benchmark"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/dataflow/concatenate_h5mu/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/dataflow/concatenate_h5mu"
+  executable: "target/nextflow/dataflow/concatenate_h5mu/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/main.nf
new file mode 100644
index 0000000..4cba42a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/main.nf
@@ -0,0 +1,4392 @@
+// concatenate_h5mu v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "concatenate_h5mu",
+  "namespace" : "dataflow",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Paths to the different samples to be concatenated.",
+          "example" : [
+            "sample_paths"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Only output concatenated objects for the provided modalities. Outputs all modalities by default.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_id",
+          "description" : "Names of the different samples that have to be concatenated.  Must be specified when using '--mode move'.\nIn this case, the ids will be used for the columns names of the dataframes registring the conflicts.\nIf specified, must be of same length as `--input`.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output location for the concatenated MuData object file.\n",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_sample_name",
+          "description" : "Name of the .obs key under which to add the sample names.",
+          "default" : [
+            "sample_id"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--other_axis_mode",
+          "description" : "How to handle the merging of other axis (var, obs, ...).\n\n  - None: keep no data\n  - same: only keep elements of the matrices which are the same in each of the samples\n  - unique: only keep elements for which there is only 1 possible value (1 value that can occur in multiple samples)\n  - first: keep the annotation from the first sample\n  - only: keep elements that show up in only one of the objects (1 unique element in only 1 sample)\n  - move: identical to 'same', but moving the conflicting values to .varm or .obsm\n",
+          "default" : [
+            "move"
+          ],
+          "required" : false,
+          "choices" : [
+            "same",
+            "unique",
+            "first",
+            "only",
+            "concat",
+            "move"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--uns_merge_mode",
+          "description" : "How to handle the merging of .uns across modalities\n  - None: keep no data\n  - same: only keep elements of the matrices which are the same in each of the samples\n  - unique: only keep elements for which there is only 1 possible value (1 value that can occur in multiple samples)\n  - first: keep the annotation from the first sample\n  - only: keep elements that show up in only one of the objects (1 unique element in only 1 sample)\n  - make_unique: identical to 'unique', but keys which are not unique are made unique by prefixing them with the sample id.\n",
+          "default" : [
+            "make_unique"
+          ],
+          "required" : false,
+          "choices" : [
+            "same",
+            "unique",
+            "first",
+            "only",
+            "make_unique"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsp_keys",
+          "description" : "List of `.obsp` keys for which block-diagonal concatenation should be performed.\nIf not provided, no `.obsp` keys will be concatenated.\nProvided keys must be present in all samples for block concatenation to be performed.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Concatenate observations from samples in several (uni- and/or multi-modal) MuData files into a single file.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data/e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data/human_brain_3k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "midcpu",
+          "highmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.13-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0",
+            "pytest-benchmark"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/dataflow/concatenate_h5mu/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/dataflow/concatenate_h5mu",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+from __future__ import annotations
+import sys
+import anndata
+import mudata as mu
+import pandas as pd
+import numpy as np
+from collections.abc import Iterable
+from multiprocessing import Pool
+from pathlib import Path
+from h5py import File as H5File
+from typing import Literal
+import shutil
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'input_id': $( if [ ! -z ${VIASH_PAR_INPUT_ID+x} ]; then echo "r'${VIASH_PAR_INPUT_ID//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_sample_name': $( if [ ! -z ${VIASH_PAR_OBS_SAMPLE_NAME+x} ]; then echo "r'${VIASH_PAR_OBS_SAMPLE_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'other_axis_mode': $( if [ ! -z ${VIASH_PAR_OTHER_AXIS_MODE+x} ]; then echo "r'${VIASH_PAR_OTHER_AXIS_MODE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'uns_merge_mode': $( if [ ! -z ${VIASH_PAR_UNS_MERGE_MODE+x} ]; then echo "r'${VIASH_PAR_UNS_MERGE_MODE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsp_keys': $( if [ ! -z ${VIASH_PAR_OBSP_KEYS+x} ]; then echo "r'${VIASH_PAR_OBSP_KEYS//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+sys.path.append(meta["resources_dir"])
+from compress_h5mu import compress_h5mu
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+
+def nunique(row):
+    unique = pd.unique(row)
+    unique_without_na = pd.core.dtypes.missing.remove_na_arraylike(unique)
+    return len(unique_without_na) > 1
+
+
+def any_row_contains_duplicate_values(n_processes: int, frame: pd.DataFrame) -> bool:
+    """
+    Check if any row contains duplicate values, that are not NA.
+    """
+    numpy_array = frame.to_numpy()
+    with Pool(n_processes) as pool:
+        is_duplicated = pool.map(nunique, iter(numpy_array))
+    return any(is_duplicated)
+
+
+def concatenate_matrices(
+    n_processes: int, matrices: dict[str, pd.DataFrame], align_to: pd.Index
+) -> tuple[
+    dict[str, pd.DataFrame], pd.DataFrame | None, dict[str, pd.core.dtypes.dtypes.Dtype]
+]:
+    """
+    Merge matrices by combining columns that have the same name.
+    Columns that contain conflicting values (e.i. the columns have different values),
+    are not merged, but instead moved to a new dataframe.
+    """
+    column_names = set(column_name for var in matrices.values() for column_name in var)
+    logger.debug("Trying to concatenate columns: %s.", ",".join(column_names))
+    if not column_names:
+        return {}, pd.DataFrame(index=align_to)
+    conflicts, concatenated_matrix = split_conflicts_and_concatenated_columns(
+        n_processes, matrices, column_names, align_to
+    )
+    concatenated_matrix = cast_to_writeable_dtype(concatenated_matrix)
+    conflicts = {
+        conflict_name: cast_to_writeable_dtype(conflict_df)
+        for conflict_name, conflict_df in conflicts.items()
+    }
+    return conflicts, concatenated_matrix
+
+
+def get_first_non_na_value_vector(df):
+    numpy_arr = df.to_numpy()
+    n_rows, n_cols = numpy_arr.shape
+    col_index = pd.isna(numpy_arr).argmin(axis=1)
+    flat_index = n_cols * np.arange(n_rows) + col_index
+    return pd.Series(numpy_arr.ravel()[flat_index], index=df.index, name=df.columns[0])
+
+
+def make_uns_keys_unique(mod_data, concatenated_data):
+    """
+    Check if the uns keys across samples are unique before adding them
+    to the final concatenated object. If a conflict occurs between the samples,
+    add the sample ID to make the key unique again.
+    """
+    all_uns_keys = {}
+    for sample_id, mod in mod_data.items():
+        for uns_key, _ in mod.uns.items():
+            all_uns_keys.setdefault(uns_key, []).append(sample_id)
+    for uns_key, samples_ids in all_uns_keys.items():
+        assert samples_ids
+        if len(samples_ids) == 1:
+            sample_id = samples_ids[0]
+            concatenated_data.uns[uns_key] = mod_data[sample_id].uns[uns_key]
+        else:
+            for sample_id in samples_ids:
+                concatenated_data.uns[f"{sample_id}_{uns_key}"] = mod_data[
+                    sample_id
+                ].uns[uns_key]
+    return concatenated_data
+
+
+def split_conflicts_and_concatenated_columns(
+    n_processes: int,
+    matrices: dict[str, pd.DataFrame],
+    column_names: Iterable[str],
+    align_to: pd.Index,
+) -> tuple[dict[str, pd.DataFrame], pd.DataFrame]:
+    """
+    Retrieve columns with the same name from a list of dataframes which are
+    identical across all the frames (ignoring NA values).
+    Columns which are not the same are regarded as 'conflicts',
+    which are stored in seperate dataframes, one per columns
+    with the same name that store conflicting values.
+    """
+    conflicts = {}
+    concatenated_matrix = []
+    for column_name in column_names:
+        columns = {
+            input_id: var[column_name]
+            for input_id, var in matrices.items()
+            if column_name in var
+        }
+        assert columns, "Some columns should have been found."
+        concatenated_columns = pd.concat(
+            columns.values(), axis=1, join="outer", sort=False
+        )
+        if any_row_contains_duplicate_values(n_processes, concatenated_columns):
+            concatenated_columns.columns = (
+                columns.keys()
+            )  # Use the sample id as column name
+            concatenated_columns = concatenated_columns.reindex(align_to, copy=False)
+            conflicts[f"conflict_{column_name}"] = concatenated_columns
+        else:
+            unique_values = get_first_non_na_value_vector(concatenated_columns)
+            concatenated_matrix.append(unique_values)
+    if not concatenated_matrix:
+        return conflicts, pd.DataFrame(index=align_to)
+    concatenated_matrix = pd.concat(
+        concatenated_matrix, join="outer", axis=1, sort=False
+    )
+    concatenated_matrix = concatenated_matrix.reindex(align_to, copy=False)
+    return conflicts, concatenated_matrix
+
+
+def cast_to_writeable_dtype(result: pd.DataFrame) -> pd.DataFrame:
+    """
+    Cast the dataframe to dtypes that can be written by mudata.
+    """
+    # dtype inferral workfs better with np.nan
+    result = result.replace({pd.NA: np.nan})
+
+    # MuData supports nullable booleans and ints
+    # ie. \\`IntegerArray\\` and \\`BooleanArray\\`
+    result = result.convert_dtypes(
+        infer_objects=True,
+        convert_integer=True,
+        convert_string=False,
+        convert_boolean=True,
+        convert_floating=False,
+    )
+
+    # Convert leftover 'object' columns to string
+    # However, na values are supported, so convert all values except NA's to string
+    object_cols = result.select_dtypes(include="object").columns.values
+    for obj_col in object_cols:
+        result[obj_col] = (
+            result[obj_col]
+            .where(result[obj_col].isna(), result[obj_col].astype(str))
+            .astype("category")
+        )
+    return result
+
+
+def split_conflicts_modalities(
+    n_processes: int, samples: dict[str, anndata.AnnData], output: anndata.AnnData
+) -> anndata.AnnData:
+    """
+    Merge .var and .obs matrices of the anndata objects. Columns are merged
+    when the values (excl NA) are the same in each of the matrices.
+    Conflicting columns are moved to a separate dataframe (one dataframe for each column,
+    containing all the corresponding column from each sample).
+    """
+    matrices_to_parse = ("var", "obs")
+    for matrix_name in matrices_to_parse:
+        matrices = {
+            sample_id: getattr(sample, matrix_name)
+            for sample_id, sample in samples.items()
+        }
+        output_index = getattr(output, matrix_name).index
+        conflicts, concatenated_matrix = concatenate_matrices(
+            n_processes, matrices, output_index
+        )
+        if concatenated_matrix.empty:
+            concatenated_matrix.index = output_index
+
+        # Even though we did not touch the varm and obsm matrices that were already present,
+        # the joining of observations might have caused a dtype change in these matrices as well
+        # so these also need to be casted to a writable dtype...
+        for multidim_name, multidim_data in getattr(output, f"{matrix_name}m").items():
+            new_data = (
+                cast_to_writeable_dtype(multidim_data)
+                if isinstance(multidim_data, pd.DataFrame)
+                else multidim_data
+            )
+            getattr(output, f"{matrix_name}m")[multidim_name] = new_data
+
+        # Write the conflicts to the output
+        for conflict_name, conflict_data in conflicts.items():
+            getattr(output, f"{matrix_name}m")[conflict_name] = conflict_data
+
+        # Set other annotation matrices in the output
+        setattr(output, matrix_name, concatenated_matrix)
+
+    return output
+
+
+def concatenate_modality(
+    n_processes: int,
+    mod: str | None,
+    input_files: Iterable[str | Path],
+    other_axis_mode: str,
+    uns_merge_mode: str,
+    input_ids: tuple[str],
+) -> anndata.AnnData:
+    concat_modes = {
+        "move": "unique",
+    }
+    other_axis_mode_to_apply = concat_modes.get(other_axis_mode, other_axis_mode)
+
+    uns_merge_modes = {"make_unique": None}
+    uns_merge_mode_to_apply = uns_merge_modes.get(uns_merge_mode, uns_merge_mode)
+
+    mod_data = {}
+    mod_indices_combined = pd.Index([])
+    for input_id, input_file in zip(input_ids, input_files):
+        if mod is not None:
+            try:
+                data = mu.read_h5ad(input_file, mod=mod)
+
+                # Remove obsp keys that are not in par["obsp_keys"]
+                obsp_keys_to_keep = par.get("obsp_keys") or []
+                obsp_keys_to_remove = set(data.obsp.keys()) - set(obsp_keys_to_keep)
+                for key in obsp_keys_to_remove:
+                    try:
+                        del data.obsp[key]
+                    except KeyError:
+                        pass
+
+                mod_data[input_id] = data
+                mod_indices_combined = mod_indices_combined.append(data.obs.index)
+
+            except KeyError as e:  # Modality does not exist for this sample, skip it
+                if (
+                    f"Unable to synchronously open object (object '{mod}' doesn't exist)"
+                    not in str(e)
+                ):
+                    raise e
+                pass
+        else:  # When mod=None, process the 'global' h5mu state
+            with H5File(input_file, "r") as input_h5:
+                if "uns" in input_h5.keys():
+                    uns_data = anndata.experimental.read_elem(input_h5["uns"])
+                    if uns_data:
+                        mod_data[input_id] = anndata.AnnData(uns=uns_data)
+
+    if not mod_indices_combined.is_unique:
+        raise ValueError("Observations are not unique across samples.")
+
+    if not mod_data:
+        return anndata.AnnData()
+
+    concatenated_data = anndata.concat(
+        mod_data.values(),
+        join="outer",
+        pairwise=True if par["obsp_keys"] else False,
+        merge=other_axis_mode_to_apply,
+        uns_merge=uns_merge_mode_to_apply,
+    )
+
+    if other_axis_mode == "move":
+        concatenated_data = split_conflicts_modalities(
+            n_processes, mod_data, concatenated_data
+        )
+
+    if uns_merge_mode == "make_unique":
+        concatenated_data = make_uns_keys_unique(mod_data, concatenated_data)
+
+    return concatenated_data
+
+
+def concatenate_modalities(
+    n_processes: int,
+    modalities: list[str],
+    input_files: Path | str,
+    other_axis_mode: str,
+    uns_merge_mode: str,
+    output_file: Path | str,
+    compression: Literal["gzip"] | Literal["lzf"],
+    input_ids: tuple[str] | None = None,
+) -> None:
+    """
+    Join the modalities together into a single multimodal sample.
+    """
+    logger.info("Concatenating samples.")
+    output_file, input_files = (
+        Path(output_file),
+        [Path(input_file) for input_file in input_files],
+    )
+    output_file_uncompressed = output_file.with_name(
+        output_file.stem + "_uncompressed.h5mu"
+    )
+    output_file_uncompressed.touch()
+    # Create empty mudata file
+    mdata = mu.MuData({modality: anndata.AnnData() for modality in modalities})
+    mdata.write(output_file_uncompressed, compression=compression)
+
+    # Use "None" for the global slots (not assigned to any modality)
+    for mod_name in modalities + [
+        None,
+    ]:
+        new_mod = concatenate_modality(
+            n_processes,
+            mod_name,
+            input_files,
+            other_axis_mode,
+            uns_merge_mode,
+            input_ids,
+        )
+        if mod_name is None:
+            if new_mod.uns:
+                with H5File(output_file_uncompressed, "r+") as open_h5mu_file:
+                    anndata.experimental.write_elem(
+                        open_h5mu_file, "uns", dict(new_mod.uns)
+                    )
+            continue
+        logger.info(
+            "Writing out modality '%s' to '%s' with compression '%s'.",
+            mod_name,
+            output_file_uncompressed,
+            compression,
+        )
+        mu.write_h5ad(output_file_uncompressed, data=new_mod, mod=mod_name)
+
+    if compression:
+        compress_h5mu(output_file_uncompressed, output_file, compression=compression)
+        output_file_uncompressed.unlink()
+    else:
+        shutil.move(output_file_uncompressed, output_file)
+
+    logger.info("Concatenation successful.")
+
+
+def main() -> None:
+    # Get a list of all possible modalities
+    mods = set()
+    for path in par["input"]:
+        try:
+            with H5File(path, "r") as f_root:
+                mods = mods | set(f_root["mod"].keys())
+        except OSError:
+            raise OSError(f"Failed to load {path}. Is it a valid h5 file?")
+
+    input_ids = None
+    if par["input_id"]:
+        input_ids: tuple[str] = tuple(i.strip() for i in par["input_id"])
+        if len(input_ids) != len(par["input"]):
+            raise ValueError(
+                "The number of sample names must match the number of sample files."
+            )
+
+        if len(set(input_ids)) != len(input_ids):
+            raise ValueError("The sample names should be unique.")
+
+    logger.info("\\\\nConcatenating data from paths:\\\\n\\\\t%s", "\\\\n\\\\t".join(par["input"]))
+
+    if par["other_axis_mode"] == "move" and not input_ids:
+        raise ValueError("--mode 'move' requires --input_ids.")
+
+    n_processes = meta["cpus"] if meta["cpus"] else 1
+
+    if par["modality"]:
+        par["modality"] = set(par["modality"])
+        if not par["modality"].issubset(mods):
+            mods_joined, input_mods_joined = ", ".join(mods), ", ".join(par["modality"])
+            raise ValueError(
+                f"One of the modalities provided ({input_mods_joined}) is not available in the input data {mods_joined}"
+            )
+        mods = par["modality"]
+
+    concatenate_modalities(
+        n_processes,
+        list(mods),
+        par["input"],
+        par["other_axis_mode"],
+        par["uns_merge_mode"],
+        par["output"],
+        par["output_compression"],
+        input_ids=input_ids,
+    )
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/dataflow/concatenate_h5mu",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "midcpu",
+    "highmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/nextflow.config
new file mode 100644
index 0000000..c6504c9
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'dataflow/concatenate_h5mu'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Concatenate observations from samples in several (uni- and/or multi-modal) MuData files into a single file.\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/nextflow_schema.json
new file mode 100644
index 0000000..53237e9
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/nextflow_schema.json
@@ -0,0 +1,118 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "concatenate_h5mu",
+  "description": "Concatenate observations from samples in several (uni- and/or multi-modal) MuData files into a single file.\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "exists": true,
+          "description": "Paths to the different samples to be concatenated.",
+          "help_text": "Type: `file`, multiple: `True`, required, direction: `input`, example: `[\"sample_paths\"]`. "
+        },
+        "modality": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Only output concatenated objects for the provided modalities",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "input_id": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Names of the different samples that have to be concatenated",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output location for the concatenated MuData object file.\n",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "obs_sample_name": {
+          "type": "string",
+          "description": "Name of the .obs key under which to add the sample names.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"sample_id\"`. ",
+          "default": "sample_id"
+        },
+        "other_axis_mode": {
+          "type": "string",
+          "description": "How to handle the merging of other axis (var, obs, ...).\n\n  - None: keep no data\n  - same: only keep elements of the matrices which are the same in each of the samples\n  - unique: only keep elements for which there is only 1 possible value (1 value that can occur in multiple samples)\n  - first: keep the annotation from the first sample\n  - only: keep elements that show up in only one of the objects (1 unique element in only 1 sample)\n  - move: identical to 'same', but moving the conflicting values to .varm or .obsm\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"move\"`, choices: ``same`, `unique`, `first`, `only`, `concat`, `move``. ",
+          "enum": [
+            "same",
+            "unique",
+            "first",
+            "only",
+            "concat",
+            "move"
+          ],
+          "default": "move"
+        },
+        "uns_merge_mode": {
+          "type": "string",
+          "description": "How to handle the merging of .uns across modalities\n  - None: keep no data\n  - same: only keep elements of the matrices which are the same in each of the samples\n  - unique: only keep elements for which there is only 1 possible value (1 value that can occur in multiple samples)\n  - first: keep the annotation from the first sample\n  - only: keep elements that show up in only one of the objects (1 unique element in only 1 sample)\n  - make_unique: identical to 'unique', but keys which are not unique are made unique by prefixing them with the sample id.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"make_unique\"`, choices: ``same`, `unique`, `first`, `only`, `make_unique``. ",
+          "enum": [
+            "same",
+            "unique",
+            "first",
+            "only",
+            "make_unique"
+          ],
+          "default": "make_unique"
+        },
+        "obsp_keys": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "List of `.obsp` keys for which block-diagonal concatenation should be performed.\nIf not provided, no `.obsp` keys will be concatenated.\nProvided keys must be present in all samples for block concatenation to be performed.\n",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/concatenate_h5mu/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/.config.vsh.yaml
new file mode 100644
index 0000000..8b39ea7
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/.config.vsh.yaml
@@ -0,0 +1,252 @@
+name: "merge"
+namespace: "dataflow"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Paths to the single-modality .h5mu files that need to be combined"
+    info: null
+    default:
+    - "sample_paths"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Path to the output file."
+    info: null
+    default:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "The compression format to be used on the output h5mu object."
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Combine one or more single-modality .h5mu files together into one .h5mu\
+  \ file.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3_filtered_feature_bc_matrix_rna.h5mu"
+- type: "file"
+  path: "pbmc_1k_protein_v3_filtered_feature_bc_matrix_prot.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "highmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/dataflow/merge/config.vsh.yml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/dataflow/merge"
+  executable: "target/nextflow/dataflow/merge/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/main.nf
new file mode 100644
index 0000000..c780d73
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/main.nf
@@ -0,0 +1,3973 @@
+// merge v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "merge",
+  "namespace" : "dataflow",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Paths to the single-modality .h5mu files that need to be combined",
+          "default" : [
+            "sample_paths"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Path to the output file.",
+          "default" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "The compression format to be used on the output h5mu object.",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Combine one or more single-modality .h5mu files together into one .h5mu file.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "../../../resources_test/merge_test_data/pbmc_1k_protein_v3_filtered_feature_bc_matrix_rna.h5mu"
+    },
+    {
+      "type" : "file",
+      "path" : "../../../resources_test/merge_test_data/pbmc_1k_protein_v3_filtered_feature_bc_matrix_prot.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "highmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/dataflow/merge/config.vsh.yml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/dataflow/merge",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+from __future__ import annotations
+import sys
+import mudata as md
+import pandas as pd
+import numpy as np
+
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+
+def main():
+    logger.info("Reading input files %s", ",".join(par["input"]))
+    input_samples = [md.read_h5mu(path) for path in par["input"]]
+
+    logger.info("Merging into single object.")
+    sample_modalities = {}
+    for input_sample in input_samples:
+        for mod_name, mod_data in input_sample.mod.items():
+            if mod_name in sample_modalities:
+                raise ValueError(
+                    f"Modality '{mod_name}' was found in more than 1 sample."
+                )
+            sample_modalities[mod_name] = mod_data
+
+    merged = md.MuData(sample_modalities)
+    merged.update()
+    for df_attr in ("var", "obs"):
+        df = getattr(merged, df_attr)
+        df = df.replace({pd.NA: np.nan}, inplace=False)
+
+        # MuData supports nullable booleans and ints
+        # ie. \\`IntegerArray\\` and \\`BooleanArray\\`
+        df = df.convert_dtypes(
+            infer_objects=True,
+            convert_integer=True,
+            convert_string=False,
+            convert_boolean=True,
+            convert_floating=False,
+        )
+
+        # pd.convert_dtypes does not convert already existing nullable dtypes
+        # to their native numpy dtypes.
+        # At this point, integer and boolean columns use a nullable dtype; and string columns
+        # might use 'object' or pd.StringDtype. This is OK.
+        # However, for example floating number columns might still use a nullable dtype
+        # from before the call to convert_dtypes. So we need to explicitly cast.
+        col_dtypes = df.select_dtypes(
+            exclude=["integer", "string", "object", "boolean"]
+        ).dtypes.to_dict()
+        numpy_dtypes = {
+            col_name: np.dtype(col_dtype.kind)
+            for col_name, col_dtype in col_dtypes.items()
+            if pd.api.types.is_extension_array_dtype(col_dtype)
+        }
+        df = df.astype(numpy_dtypes)
+
+        # Convert leftover 'object' columns to string
+        object_cols = df.select_dtypes(include="object").columns.values
+        for obj_col in object_cols:
+            df[obj_col] = df[obj_col].astype(str).astype("category")
+
+        setattr(merged, df_attr, df)
+
+    merged.write_h5mu(par["output"], compression=par["output_compression"])
+    logger.info("Finished")
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/dataflow/merge",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "singlecpu",
+    "highmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/nextflow.config
new file mode 100644
index 0000000..d1b44ff
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'dataflow/merge'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Combine one or more single-modality .h5mu files together into one .h5mu file.\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/nextflow_schema.json
new file mode 100644
index 0000000..2f20aa0
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/nextflow_schema.json
@@ -0,0 +1,64 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "merge",
+  "description": "Combine one or more single-modality .h5mu files together into one .h5mu file.\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "exists": true,
+          "description": "Paths to the single-modality .h5mu files that need to be combined",
+          "help_text": "Type: `file`, multiple: `True`, required, default: `[\"sample_paths\"]`, direction: `input`. ",
+          "default": [
+            "sample_paths"
+          ]
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to the output file.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"output.h5mu\"`, direction: `output`. ",
+          "default": "output.h5mu"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "The compression format to be used on the output h5mu object.",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/merge/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/.config.vsh.yaml
new file mode 100644
index 0000000..d085ecf
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/.config.vsh.yaml
@@ -0,0 +1,279 @@
+name: "split_modalities"
+namespace: "dataflow"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "contributor"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to a single .h5mu file."
+    info: null
+    default:
+    - "sample_path"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output directory containing multiple h5mu files."
+    info: null
+    example:
+    - "/path/to/output"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_types"
+    description: "A csv containing the base filename and modality type per output\
+      \ file."
+    info: null
+    example:
+    - "types.csv"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Split the modalities from a single .h5mu multimodal sample into seperate\
+  \ .h5mu files. \n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "lowmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/dataflow/split_modalities/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/dataflow/split_modalities"
+  executable: "target/nextflow/dataflow/split_modalities/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/main.nf
new file mode 100644
index 0000000..208430d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/main.nf
@@ -0,0 +1,3991 @@
+// split_modalities v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+//  * Robrecht Cannoodt (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "split_modalities",
+  "namespace" : "dataflow",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to a single .h5mu file.",
+          "default" : [
+            "sample_path"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output directory containing multiple h5mu files.",
+          "example" : [
+            "/path/to/output"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_types",
+          "description" : "A csv containing the base filename and modality type per output file.",
+          "example" : [
+            "types.csv"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Split the modalities from a single .h5mu multimodal sample into seperate .h5mu files. \n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "lowmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/dataflow/split_modalities/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/dataflow/split_modalities",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+from __future__ import annotations
+import sys
+import mudata as md
+from pathlib import Path
+import pandas as pd
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_types': $( if [ ! -z ${VIASH_PAR_OUTPUT_TYPES+x} ]; then echo "r'${VIASH_PAR_OUTPUT_TYPES//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+
+def main() -> None:
+    output_dir = Path(par["output"])
+    logger.info("Creating output directory '%s' if it does not exist", output_dir)
+    if not output_dir.is_dir():
+        logger.info("Creating %s", output_dir)
+        output_dir.mkdir(parents=True)
+
+    logger.info("Reading input file '%s'", par["input"])
+    input_file = Path(par["input"].strip())
+    sample = md.read_h5mu(input_file)
+
+    logger.info("Creating output types CSV.")
+    modalities = list(sample.mod.keys())
+
+    logger.info("Found the following modalities:\\\\n%s", "\\\\n".join(modalities))
+    names = {mod_name: f"{input_file.stem}_{mod_name}.h5mu" for mod_name in modalities}
+    output_files = list(names.values())
+    logger.info(
+        "Will be creating the following output .h5mu files:\\\\n%s",
+        "\\\\n".join(output_files),
+    )
+    df = pd.DataFrame({"name": modalities, "filename": output_files})
+    logger.info("Writing output_types CSV file to '%s'.", par["output_types"])
+    df.to_csv(par["output_types"], index=False)
+
+    logger.info("Splitting input file into unimodal output files.")
+    for mod_name, mod in sample.mod.items():
+        logger.info("Processing modality '%s'", mod_name)
+        new_sample = md.MuData({mod_name: mod})
+        logger.info(
+            "Writing to '%s', with compression '%s'",
+            names[mod_name],
+            par["output_compression"],
+        )
+        new_sample.write_h5mu(
+            output_dir / names[mod_name], compression=par["output_compression"]
+        )
+        logger.info("Done writing output file.")
+    logger.info("Finished")
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/dataflow/split_modalities",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "singlecpu",
+    "lowmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/nextflow.config
new file mode 100644
index 0000000..03c027c
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'dataflow/split_modalities'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Split the modalities from a single .h5mu multimodal sample into seperate .h5mu files. \n'
+  author = 'Dries Schaumont, Robrecht Cannoodt'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/nextflow_schema.json
new file mode 100644
index 0000000..7e4a6d1
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/nextflow_schema.json
@@ -0,0 +1,66 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "split_modalities",
+  "description": "Split the modalities from a single .h5mu multimodal sample into seperate .h5mu files. \n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to a single .h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"sample_path\"`, direction: `input`. ",
+          "default": "sample_path"
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output directory containing multiple h5mu files.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output\"`, direction: `output`, example: `\"/path/to/output\"`. ",
+          "default": "$id.$key.output"
+        },
+        "output_types": {
+          "type": "string",
+          "format": "path",
+          "description": "A csv containing the base filename and modality type per output file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output_types.csv\"`, direction: `output`, example: `\"types.csv\"`. ",
+          "default": "$id.$key.output_types.csv"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dataflow/split_modalities/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/.config.vsh.yaml
new file mode 100644
index 0000000..a2568e2
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/.config.vsh.yaml
@@ -0,0 +1,324 @@
+name: "pca"
+namespace: "dimred"
+version: "v4.0.4"
+authors:
+- name: "Dries De Maeyer"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Use specified layer for expression values instead of the .X object\
+      \ from the modality."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_input"
+    description: "Column name in .var matrix that will be used to select which genes\
+      \ to run the PCA on."
+    info: null
+    example:
+    - "filter_with_hvg"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_output"
+    description: "In which .obsm slot to store the resulting embedding."
+    info: null
+    default:
+    - "X_pca"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--varm_output"
+    description: "In which .varm slot to store the resulting loadings matrix."
+    info: null
+    default:
+    - "pca_loadings"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--uns_output"
+    description: "In which .uns slot to store the resulting variance objects."
+    info: null
+    default:
+    - "pca_variance"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--num_components"
+    description: "Number of principal components to compute. Defaults to 50, or 1\
+      \ - minimum dimension size of selected representation."
+    info: null
+    example:
+    - 25
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--overwrite"
+    description: "Allow overwriting .obsm, .varm and .uns slots."
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Computes PCA coordinates, loadings and variance decomposition. Uses\
+  \ the implementation of scikit-learn [Pedregosa11].\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "highcpu"
+    - "highmem"
+    - "middisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/dimred/pca/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/dimred/pca"
+  executable: "target/nextflow/dimred/pca/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/main.nf
new file mode 100644
index 0000000..3bc0c07
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/main.nf
@@ -0,0 +1,4060 @@
+// pca v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries De Maeyer (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "pca",
+  "namespace" : "dimred",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Use specified layer for expression values instead of the .X object from the modality.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_input",
+          "description" : "Column name in .var matrix that will be used to select which genes to run the PCA on.",
+          "example" : [
+            "filter_with_hvg"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_output",
+          "description" : "In which .obsm slot to store the resulting embedding.",
+          "default" : [
+            "X_pca"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--varm_output",
+          "description" : "In which .varm slot to store the resulting loadings matrix.",
+          "default" : [
+            "pca_loadings"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--uns_output",
+          "description" : "In which .uns slot to store the resulting variance objects.",
+          "default" : [
+            "pca_variance"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--num_components",
+          "description" : "Number of principal components to compute. Defaults to 50, or 1 - minimum dimension size of selected representation.",
+          "example" : [
+            25
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--overwrite",
+          "description" : "Allow overwriting .obsm, .varm and .uns slots.",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Computes PCA coordinates, loadings and variance decomposition. Uses the implementation of scikit-learn [Pedregosa11].\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "highcpu",
+          "highmem",
+          "middisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/dimred/pca/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/dimred/pca",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import scanpy as sc
+import mudata as mu
+import sys
+from anndata import AnnData
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'layer': $( if [ ! -z ${VIASH_PAR_LAYER+x} ]; then echo "r'${VIASH_PAR_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_input': $( if [ ! -z ${VIASH_PAR_VAR_INPUT+x} ]; then echo "r'${VIASH_PAR_VAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_output': $( if [ ! -z ${VIASH_PAR_OBSM_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OBSM_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'varm_output': $( if [ ! -z ${VIASH_PAR_VARM_OUTPUT+x} ]; then echo "r'${VIASH_PAR_VARM_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'uns_output': $( if [ ! -z ${VIASH_PAR_UNS_OUTPUT+x} ]; then echo "r'${VIASH_PAR_UNS_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'num_components': $( if [ ! -z ${VIASH_PAR_NUM_COMPONENTS+x} ]; then echo "int(r'${VIASH_PAR_NUM_COMPONENTS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'overwrite': $( if [ ! -z ${VIASH_PAR_OVERWRITE+x} ]; then echo "r'${VIASH_PAR_OVERWRITE//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+logger.info("Reading %s, modality %s", par["input"], par["modality"])
+data = mu.read_h5ad(par["input"], mod=par["modality"])
+
+logger.info("Computing PCA components for modality '%s'", par["modality"])
+if par["layer"] and par["layer"] not in data.layers:
+    raise ValueError(f"{par['layer']} was not found in modality {par['modality']}.")
+layer = data.X if not par["layer"] else data.layers[par["layer"]]
+adata_input_layer = AnnData(layer)
+adata_input_layer.var.index = data.var.index
+
+use_highly_variable = False
+if par["var_input"]:
+    if par["var_input"] not in data.var.columns:
+        raise ValueError(
+            f"Requested to use .var column {par['var_input']} "
+            "as a selection of genes to run the PCA on, "
+            f"but the column is not available for modality {par['modality']}"
+        )
+    use_highly_variable = True
+    adata_input_layer.var["highly_variable"] = data.var[par["var_input"]]
+
+# run pca
+output_adata = sc.tl.pca(
+    adata_input_layer,
+    n_comps=par["num_components"],
+    copy=True,
+    use_highly_variable=use_highly_variable,
+)
+
+# store output in specific objects
+
+check_exist_dict = {
+    "obsm_output": ("obs"),
+    "varm_output": ("varm"),
+    "uns_output": ("uns"),
+}
+for parameter_name, field in check_exist_dict.items():
+    if par[parameter_name] in getattr(data, field):
+        if not par["overwrite"]:
+            raise ValueError(
+                f"Requested to create field {par[parameter_name]} in .{field} "
+                f"for modality {par['modality']}, but field already exists."
+            )
+        del getattr(data, field)[par[parameter_name]]
+
+data.obsm[par["obsm_output"]] = output_adata.obsm["X_pca"]
+data.varm[par["varm_output"]] = output_adata.varm["PCs"]
+data.uns[par["uns_output"]] = {
+    "variance": output_adata.uns["pca"]["variance"],
+    "variance_ratio": output_adata.uns["pca"]["variance_ratio"],
+}
+
+
+logger.info(
+    "Writing to %s with compression %s.", par["output"], par["output_compression"]
+)
+write_h5ad_to_h5mu_with_compression(
+    par["output"], par["input"], par["modality"], data, par["output_compression"]
+)
+
+logger.info("Finished")
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/dimred/pca",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "highcpu",
+    "highmem",
+    "middisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/nextflow.config
new file mode 100644
index 0000000..ac1b91f
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'dimred/pca'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Computes PCA coordinates, loadings and variance decomposition. Uses the implementation of scikit-learn [Pedregosa11].\n'
+  author = 'Dries De Maeyer'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/nextflow_schema.json
new file mode 100644
index 0000000..c9eb093
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/nextflow_schema.json
@@ -0,0 +1,103 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "pca",
+  "description": "Computes PCA coordinates, loadings and variance decomposition. Uses the implementation of scikit-learn [Pedregosa11].\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "layer": {
+          "type": "string",
+          "description": "Use specified layer for expression values instead of the .X object from the modality.",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "var_input": {
+          "type": "string",
+          "description": "Column name in .var matrix that will be used to select which genes to run the PCA on.",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"filter_with_hvg\"`. "
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "obsm_output": {
+          "type": "string",
+          "description": "In which .obsm slot to store the resulting embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_pca\"`. ",
+          "default": "X_pca"
+        },
+        "varm_output": {
+          "type": "string",
+          "description": "In which .varm slot to store the resulting loadings matrix.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"pca_loadings\"`. ",
+          "default": "pca_loadings"
+        },
+        "uns_output": {
+          "type": "string",
+          "description": "In which .uns slot to store the resulting variance objects.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"pca_variance\"`. ",
+          "default": "pca_variance"
+        },
+        "num_components": {
+          "type": "integer",
+          "description": "Number of principal components to compute",
+          "help_text": "Type: `integer`, multiple: `False`, example: `25`. "
+        },
+        "overwrite": {
+          "type": "boolean",
+          "description": "Allow overwriting .obsm, .varm and .uns slots.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/pca/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/.config.vsh.yaml
new file mode 100644
index 0000000..72badb6
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/.config.vsh.yaml
@@ -0,0 +1,378 @@
+name: "umap"
+namespace: "dimred"
+version: "v4.0.4"
+authors:
+- name: "Dries De Maeyer"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--uns_neighbors"
+    description: "The `.uns` neighbors slot as output by the `find_neighbors` component."
+    info: null
+    default:
+    - "neighbors"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_output"
+    description: "The pre/postfix under which to store the UMAP results."
+    info: null
+    default:
+    - "umap"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Arguments"
+  arguments:
+  - type: "double"
+    name: "--min_dist"
+    description: "The effective minimum distance between embedded points. Smaller\
+      \ values will result in a more clustered/clumped embedding where nearby points\
+      \ on the manifold are drawn closer together, while larger values will result\
+      \ on a more even dispersal of points. The value should be set relative to the\
+      \ spread value, which determines the scale at which embedded points will be\
+      \ spread out."
+    info: null
+    default:
+    - 0.5
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--spread"
+    description: "The effective scale of embedded points. In combination with `min_dist`\
+      \ this determines how clustered/clumped the embedded points are."
+    info: null
+    default:
+    - 1.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--num_components"
+    description: "The number of dimensions of the embedding."
+    info: null
+    default:
+    - 2
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_iter"
+    description: "The number of iterations (epochs) of the optimization. Called `n_epochs`\
+      \ in the original UMAP. Default is set to 500 if neighbors['connectivities'].shape[0]\
+      \ <= 10000, else 200."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--alpha"
+    description: "The initial learning rate for the embedding optimization."
+    info: null
+    default:
+    - 1.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--gamma"
+    description: "Weighting applied to negative samples in low dimensional embedding\
+      \ optimization. Values higher than one will result in greater weight being given\
+      \ to negative samples."
+    info: null
+    default:
+    - 1.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--negative_sample_rate"
+    description: "The number of negative edge/1-simplex samples to use per positive\
+      \ edge/1-simplex sample in optimizing the low dimensional embedding."
+    info: null
+    default:
+    - 5
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--init_pos"
+    description: "How to initialize the low dimensional embedding. Called `init` in\
+      \ the original UMAP. Options are:\n  \n* Any key from `.obsm`\n* `'paga'`: positions\
+      \ from `paga()`\n* `'spectral'`: use a spectral embedding of the graph\n* `'random'`:\
+      \ assign initial embedding positions at random.\n"
+    info: null
+    default:
+    - "spectral"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "UMAP (Uniform Manifold Approximation and Projection) is a manifold learning\
+  \ technique suitable for visualizing high-dimensional data. Besides tending to be\
+  \ faster than tSNE, it optimizes the embedding such that it best reflects the topology\
+  \ of the data, which we represent throughout Scanpy using a neighborhood graph.\
+  \ tSNE, by contrast, optimizes the distribution of nearest-neighbor distances in\
+  \ the embedding such that these best match the distribution of distances in the\
+  \ high-dimensional space. We use the implementation of umap-learn [McInnes18]. For\
+  \ a few comparisons of UMAP with tSNE, see this preprint.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "highcpu"
+    - "midmem"
+    - "middisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/dimred/umap/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/dimred/umap"
+  executable: "target/nextflow/dimred/umap/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/main.nf
new file mode 100644
index 0000000..9a06d7c
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/main.nf
@@ -0,0 +1,4098 @@
+// umap v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries De Maeyer (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "umap",
+  "namespace" : "dimred",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--uns_neighbors",
+          "description" : "The `.uns` neighbors slot as output by the `find_neighbors` component.",
+          "default" : [
+            "neighbors"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_output",
+          "description" : "The pre/postfix under which to store the UMAP results.",
+          "default" : [
+            "umap"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "double",
+          "name" : "--min_dist",
+          "description" : "The effective minimum distance between embedded points. Smaller values will result in a more clustered/clumped embedding where nearby points on the manifold are drawn closer together, while larger values will result on a more even dispersal of points. The value should be set relative to the spread value, which determines the scale at which embedded points will be spread out.",
+          "default" : [
+            0.5
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--spread",
+          "description" : "The effective scale of embedded points. In combination with `min_dist` this determines how clustered/clumped the embedded points are.",
+          "default" : [
+            1.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--num_components",
+          "description" : "The number of dimensions of the embedding.",
+          "default" : [
+            2
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_iter",
+          "description" : "The number of iterations (epochs) of the optimization. Called `n_epochs` in the original UMAP. Default is set to 500 if neighbors['connectivities'].shape[0] <= 10000, else 200.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--alpha",
+          "description" : "The initial learning rate for the embedding optimization.",
+          "default" : [
+            1.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--gamma",
+          "description" : "Weighting applied to negative samples in low dimensional embedding optimization. Values higher than one will result in greater weight being given to negative samples.",
+          "default" : [
+            1.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--negative_sample_rate",
+          "description" : "The number of negative edge/1-simplex samples to use per positive edge/1-simplex sample in optimizing the low dimensional embedding.",
+          "default" : [
+            5
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--init_pos",
+          "description" : "How to initialize the low dimensional embedding. Called `init` in the original UMAP. Options are:\n  \n* Any key from `.obsm`\n* `'paga'`: positions from `paga()`\n* `'spectral'`: use a spectral embedding of the graph\n* `'random'`: assign initial embedding positions at random.\n",
+          "default" : [
+            "spectral"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "UMAP (Uniform Manifold Approximation and Projection) is a manifold learning technique suitable for visualizing high-dimensional data. Besides tending to be faster than tSNE, it optimizes the embedding such that it best reflects the topology of the data, which we represent throughout Scanpy using a neighborhood graph. tSNE, by contrast, optimizes the distribution of nearest-neighbor distances in the embedding such that these best match the distribution of distances in the high-dimensional space. We use the implementation of umap-learn [McInnes18]. For a few comparisons of UMAP with tSNE, see this preprint.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "highcpu",
+          "midmem",
+          "middisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/dimred/umap/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/dimred/umap",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import scanpy as sc
+import mudata as mu
+import sys
+import anndata as ad
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'uns_neighbors': $( if [ ! -z ${VIASH_PAR_UNS_NEIGHBORS+x} ]; then echo "r'${VIASH_PAR_UNS_NEIGHBORS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_output': $( if [ ! -z ${VIASH_PAR_OBSM_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OBSM_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'min_dist': $( if [ ! -z ${VIASH_PAR_MIN_DIST+x} ]; then echo "float(r'${VIASH_PAR_MIN_DIST//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'spread': $( if [ ! -z ${VIASH_PAR_SPREAD+x} ]; then echo "float(r'${VIASH_PAR_SPREAD//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'num_components': $( if [ ! -z ${VIASH_PAR_NUM_COMPONENTS+x} ]; then echo "int(r'${VIASH_PAR_NUM_COMPONENTS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'max_iter': $( if [ ! -z ${VIASH_PAR_MAX_ITER+x} ]; then echo "int(r'${VIASH_PAR_MAX_ITER//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'alpha': $( if [ ! -z ${VIASH_PAR_ALPHA+x} ]; then echo "float(r'${VIASH_PAR_ALPHA//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'gamma': $( if [ ! -z ${VIASH_PAR_GAMMA+x} ]; then echo "float(r'${VIASH_PAR_GAMMA//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'negative_sample_rate': $( if [ ! -z ${VIASH_PAR_NEGATIVE_SAMPLE_RATE+x} ]; then echo "int(r'${VIASH_PAR_NEGATIVE_SAMPLE_RATE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'init_pos': $( if [ ! -z ${VIASH_PAR_INIT_POS+x} ]; then echo "r'${VIASH_PAR_INIT_POS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+logger.info("Reading %s, modality %s", par["input"], par["modality"])
+data = mu.read_h5ad(par["input"], mod=par["modality"])
+
+logger.info("Computing UMAP for modality '%s'", par["modality"])
+
+if par["uns_neighbors"] not in data.uns:
+    raise ValueError(
+        f"'{par['uns_neighbors']}' was not found in .mod['{par['modality']}'].uns."
+    )
+
+# create temporary AnnData
+# ... because sc.tl.umap doesn't allow to choose
+# the obsm output slot
+# ... also we can see scanpy is a data format dependency hell
+neigh_key = par["uns_neighbors"]
+temp_uns = {neigh_key: data.uns[neigh_key]}
+conn_key = temp_uns[neigh_key]["connectivities_key"]
+dist_key = temp_uns[neigh_key]["distances_key"]
+temp_obsp = {
+    conn_key: data.obsp[conn_key],
+    dist_key: data.obsp[dist_key],
+}
+pca_key = temp_uns[neigh_key]["params"]["use_rep"]
+temp_obsm = {pca_key: data.obsm[pca_key]}
+
+temp_adata = ad.AnnData(obsm=temp_obsm, obsp=temp_obsp, uns=temp_uns, shape=data.shape)
+
+sc.tl.umap(
+    temp_adata,
+    min_dist=par["min_dist"],
+    spread=par["spread"],
+    n_components=par["num_components"],
+    maxiter=par["max_iter"],
+    alpha=par["alpha"],
+    gamma=par["gamma"],
+    negative_sample_rate=par["negative_sample_rate"],
+    init_pos=par["init_pos"],
+    neighbors_key=neigh_key,
+)
+
+data.obsm[par["obsm_output"]] = temp_adata.obsm["X_umap"]
+
+logger.info("Writing to %s, compression %s", par["output"], par["output_compression"])
+write_h5ad_to_h5mu_with_compression(
+    par["output"], par["input"], par["modality"], data, par["output_compression"]
+)
+logger.info("Finished")
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/dimred/umap",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "highcpu",
+    "midmem",
+    "middisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/nextflow.config
new file mode 100644
index 0000000..97105b4
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'dimred/umap'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'UMAP (Uniform Manifold Approximation and Projection) is a manifold learning technique suitable for visualizing high-dimensional data. Besides tending to be faster than tSNE, it optimizes the embedding such that it best reflects the topology of the data, which we represent throughout Scanpy using a neighborhood graph. tSNE, by contrast, optimizes the distribution of nearest-neighbor distances in the embedding such that these best match the distribution of distances in the high-dimensional space. We use the implementation of umap-learn [McInnes18]. For a few comparisons of UMAP with tSNE, see this preprint.\n'
+  author = 'Dries De Maeyer'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/nextflow_schema.json
new file mode 100644
index 0000000..d715953
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/nextflow_schema.json
@@ -0,0 +1,143 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "umap",
+  "description": "UMAP (Uniform Manifold Approximation and Projection) is a manifold learning technique suitable for visualizing high-dimensional data. Besides tending to be faster than tSNE, it optimizes the embedding such that it best reflects the topology of the data, which we represent throughout Scanpy using a neighborhood graph. tSNE, by contrast, optimizes the distribution of nearest-neighbor distances in the embedding such that these best match the distribution of distances in the high-dimensional space. We use the implementation of umap-learn [McInnes18]. For a few comparisons of UMAP with tSNE, see this preprint.\n",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "uns_neighbors": {
+          "type": "string",
+          "description": "The `.uns` neighbors slot as output by the `find_neighbors` component.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"neighbors\"`. ",
+          "default": "neighbors"
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "obsm_output": {
+          "type": "string",
+          "description": "The pre/postfix under which to store the UMAP results.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"umap\"`. ",
+          "default": "umap"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "min_dist": {
+          "type": "number",
+          "description": "The effective minimum distance between embedded points",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.5`. ",
+          "default": 0.5
+        },
+        "spread": {
+          "type": "number",
+          "description": "The effective scale of embedded points",
+          "help_text": "Type: `double`, multiple: `False`, default: `1.0`. ",
+          "default": 1.0
+        },
+        "num_components": {
+          "type": "integer",
+          "description": "The number of dimensions of the embedding.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `2`. ",
+          "default": 2
+        },
+        "max_iter": {
+          "type": "integer",
+          "description": "The number of iterations (epochs) of the optimization",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "alpha": {
+          "type": "number",
+          "description": "The initial learning rate for the embedding optimization.",
+          "help_text": "Type: `double`, multiple: `False`, default: `1.0`. ",
+          "default": 1.0
+        },
+        "gamma": {
+          "type": "number",
+          "description": "Weighting applied to negative samples in low dimensional embedding optimization",
+          "help_text": "Type: `double`, multiple: `False`, default: `1.0`. ",
+          "default": 1.0
+        },
+        "negative_sample_rate": {
+          "type": "integer",
+          "description": "The number of negative edge/1-simplex samples to use per positive edge/1-simplex sample in optimizing the low dimensional embedding.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `5`. ",
+          "default": 5
+        },
+        "init_pos": {
+          "type": "string",
+          "description": "How to initialize the low dimensional embedding",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"spectral\"`. ",
+          "default": "spectral"
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/dimred/umap/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/.config.vsh.yaml
new file mode 100644
index 0000000..05dfa1b
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/.config.vsh.yaml
@@ -0,0 +1,429 @@
+name: "highly_variable_features_scanpy"
+namespace: "feature_annotation"
+version: "v4.0.4"
+authors:
+- name: "Dries De Maeyer"
+  roles:
+  - "contributor"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "maintainer"
+  - "contributor"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "use adata.layers[layer] for expression values instead of adata.X."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_input"
+    description: "If specified, use boolean array in adata.var[var_input] to calculate\
+      \ hvg on subset of vars.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    description: "Output h5mu file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_filter"
+    description: "In which .var slot to store a boolean array corresponding to which\
+      \ observations should be filtered out."
+    info: null
+    default:
+    - "filter_with_hvg"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--varm_name"
+    description: "In which .varm slot to store additional metadata."
+    info: null
+    default:
+    - "hvg"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--flavor"
+    description: "Choose the flavor for identifying highly variable features. For\
+      \ the dispersion based methods\nin their default workflows, Seurat passes the\
+      \ cutoffs whereas Cell Ranger passes n_top_features.\n"
+    info: null
+    default:
+    - "seurat"
+    required: false
+    choices:
+    - "seurat"
+    - "cell_ranger"
+    - "seurat_v3"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--n_top_features"
+    description: "Number of highly-variable features to keep. Mandatory if flavor='seurat_v3'."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--min_mean"
+    description: "If n_top_features is defined, this and all other cutoffs for the\
+      \ means and the normalized dispersions are ignored. Ignored if flavor='seurat_v3'."
+    info: null
+    default:
+    - 0.0125
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--max_mean"
+    description: "If n_top_features is defined, this and all other cutoffs for the\
+      \ means and the normalized dispersions are ignored. Ignored if flavor='seurat_v3'."
+    info: null
+    default:
+    - 3.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--min_disp"
+    description: "If n_top_features is defined, this and all other cutoffs for the\
+      \ means and the normalized dispersions are ignored. Ignored if flavor='seurat_v3'."
+    info: null
+    default:
+    - 0.5
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--max_disp"
+    description: "If n_top_features is defined, this and all other cutoffs for the\
+      \ means and the normalized dispersions are ignored. Ignored if flavor='seurat_v3'.\
+      \ Default is +inf."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--span"
+    description: "The fraction of the data (cells) used when estimating the variance\
+      \ in the loess model fit if flavor='seurat_v3'."
+    info: null
+    default:
+    - 0.3
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--n_bins"
+    description: "Number of bins for binning the mean feature expression. Normalization\
+      \ is done with respect to each bin. If just a single feature falls into a bin,\
+      \ the normalized dispersion is artificially set to 1."
+    info: null
+    default:
+    - 20
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_batch_key"
+    description: "If specified, highly-variable features are selected within each\
+      \ batch separately and merged. This simple \nprocess avoids the selection of\
+      \ batch-specific features and acts as a lightweight batch correction method.\
+      \ \nFor all flavors, features are first sorted by how many batches they are\
+      \ a HVG. For dispersion-based flavors \nties are broken by normalized dispersion.\
+      \ If flavor = 'seurat_v3', ties are broken by the median (across\nbatches) rank\
+      \ based on within-batch normalized variance.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "subset_vars.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Annotate highly variable features [Satija15] [Zheng17] [Stuart19].\n\
+  \nExpects logarithmized data, except when flavor='seurat_v3' in which count data\
+  \ is expected.\n\nDepending on flavor, this reproduces the R-implementations of\
+  \ Seurat [Satija15], Cell Ranger [Zheng17], and Seurat v3 [Stuart19].\n\nFor the\
+  \ dispersion-based methods ([Satija15] and [Zheng17]), the normalized dispersion\
+  \ is obtained by scaling with the mean and standard deviation of the dispersions\
+  \ for features falling into a given bin for mean expression of features. This means\
+  \ that for each bin of mean expression, highly variable features are selected.\n\
+  \nFor [Stuart19], a normalized variance for each feature is computed. First, the\
+  \ data are standardized (i.e., z-score normalization per feature) with a regularized\
+  \ standard deviation. Next, the normalized variance is computed as the variance\
+  \ of each feature after the transformation. Features are ranked by the normalized\
+  \ variance.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "lowmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    - "scikit-misc"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/feature_annotation/highly_variable_features_scanpy/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/feature_annotation/highly_variable_features_scanpy"
+  executable: "target/nextflow/feature_annotation/highly_variable_features_scanpy/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/main.nf
new file mode 100644
index 0000000..09c37b3
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/main.nf
@@ -0,0 +1,4236 @@
+// highly_variable_features_scanpy v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries De Maeyer (contributor)
+//  * Robrecht Cannoodt (maintainer, contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "highly_variable_features_scanpy",
+  "namespace" : "feature_annotation",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "maintainer",
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "use adata.layers[layer] for expression values instead of adata.X.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_input",
+          "description" : "If specified, use boolean array in adata.var[var_input] to calculate hvg on subset of vars.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Output h5mu file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_filter",
+          "description" : "In which .var slot to store a boolean array corresponding to which observations should be filtered out.",
+          "default" : [
+            "filter_with_hvg"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--varm_name",
+          "description" : "In which .varm slot to store additional metadata.",
+          "default" : [
+            "hvg"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--flavor",
+          "description" : "Choose the flavor for identifying highly variable features. For the dispersion based methods\nin their default workflows, Seurat passes the cutoffs whereas Cell Ranger passes n_top_features.\n",
+          "default" : [
+            "seurat"
+          ],
+          "required" : false,
+          "choices" : [
+            "seurat",
+            "cell_ranger",
+            "seurat_v3"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--n_top_features",
+          "description" : "Number of highly-variable features to keep. Mandatory if flavor='seurat_v3'.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--min_mean",
+          "description" : "If n_top_features is defined, this and all other cutoffs for the means and the normalized dispersions are ignored. Ignored if flavor='seurat_v3'.",
+          "default" : [
+            0.0125
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--max_mean",
+          "description" : "If n_top_features is defined, this and all other cutoffs for the means and the normalized dispersions are ignored. Ignored if flavor='seurat_v3'.",
+          "default" : [
+            3.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--min_disp",
+          "description" : "If n_top_features is defined, this and all other cutoffs for the means and the normalized dispersions are ignored. Ignored if flavor='seurat_v3'.",
+          "default" : [
+            0.5
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--max_disp",
+          "description" : "If n_top_features is defined, this and all other cutoffs for the means and the normalized dispersions are ignored. Ignored if flavor='seurat_v3'. Default is +inf.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--span",
+          "description" : "The fraction of the data (cells) used when estimating the variance in the loess model fit if flavor='seurat_v3'.",
+          "default" : [
+            0.3
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--n_bins",
+          "description" : "Number of bins for binning the mean feature expression. Normalization is done with respect to each bin. If just a single feature falls into a bin, the normalized dispersion is artificially set to 1.",
+          "default" : [
+            20
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_batch_key",
+          "description" : "If specified, highly-variable features are selected within each batch separately and merged. This simple \nprocess avoids the selection of batch-specific features and acts as a lightweight batch correction method. \nFor all flavors, features are first sorted by how many batches they are a HVG. For dispersion-based flavors \nties are broken by normalized dispersion. If flavor = 'seurat_v3', ties are broken by the median (across\nbatches) rank based on within-batch normalized variance.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/subset_vars.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Annotate highly variable features [Satija15] [Zheng17] [Stuart19].\n\nExpects logarithmized data, except when flavor='seurat_v3' in which count data is expected.\n\nDepending on flavor, this reproduces the R-implementations of Seurat [Satija15], Cell Ranger [Zheng17], and Seurat v3 [Stuart19].\n\nFor the dispersion-based methods ([Satija15] and [Zheng17]), the normalized dispersion is obtained by scaling with the mean and standard deviation of the dispersions for features falling into a given bin for mean expression of features. This means that for each bin of mean expression, highly variable features are selected.\n\nFor [Stuart19], a normalized variance for each feature is computed. First, the data are standardized (i.e., z-score normalization per feature) with a regularized standard deviation. Next, the normalized variance is computed as the variance of each feature after the transformation. Features are ranked by the normalized variance.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "lowmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4",
+            "scikit-misc"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/feature_annotation/highly_variable_features_scanpy/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/feature_annotation/highly_variable_features_scanpy",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import scanpy as sc
+import mudata as mu
+import anndata as ad
+import pandas as pd
+import sys
+import re
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'layer': $( if [ ! -z ${VIASH_PAR_LAYER+x} ]; then echo "r'${VIASH_PAR_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_input': $( if [ ! -z ${VIASH_PAR_VAR_INPUT+x} ]; then echo "r'${VIASH_PAR_VAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_name_filter': $( if [ ! -z ${VIASH_PAR_VAR_NAME_FILTER+x} ]; then echo "r'${VIASH_PAR_VAR_NAME_FILTER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'varm_name': $( if [ ! -z ${VIASH_PAR_VARM_NAME+x} ]; then echo "r'${VIASH_PAR_VARM_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'flavor': $( if [ ! -z ${VIASH_PAR_FLAVOR+x} ]; then echo "r'${VIASH_PAR_FLAVOR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'n_top_features': $( if [ ! -z ${VIASH_PAR_N_TOP_FEATURES+x} ]; then echo "int(r'${VIASH_PAR_N_TOP_FEATURES//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'min_mean': $( if [ ! -z ${VIASH_PAR_MIN_MEAN+x} ]; then echo "float(r'${VIASH_PAR_MIN_MEAN//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'max_mean': $( if [ ! -z ${VIASH_PAR_MAX_MEAN+x} ]; then echo "float(r'${VIASH_PAR_MAX_MEAN//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'min_disp': $( if [ ! -z ${VIASH_PAR_MIN_DISP+x} ]; then echo "float(r'${VIASH_PAR_MIN_DISP//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'max_disp': $( if [ ! -z ${VIASH_PAR_MAX_DISP+x} ]; then echo "float(r'${VIASH_PAR_MAX_DISP//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'span': $( if [ ! -z ${VIASH_PAR_SPAN+x} ]; then echo "float(r'${VIASH_PAR_SPAN//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'n_bins': $( if [ ! -z ${VIASH_PAR_N_BINS+x} ]; then echo "int(r'${VIASH_PAR_N_BINS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'obs_batch_key': $( if [ ! -z ${VIASH_PAR_OBS_BATCH_KEY+x} ]; then echo "r'${VIASH_PAR_OBS_BATCH_KEY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from subset_vars import subset_vars
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+modality_name = par["modality"]
+data = mu.read_h5ad(par["input"], mod=modality_name)
+assert data.var_names.is_unique, "Expected var_names of input modality to be unique."
+
+logger.info("Processing modality '%s'", modality_name)
+
+if par["layer"] and par["layer"] not in data.layers:
+    raise ValueError(
+        f"Layer '{par['layer']}' not found in layers for modality '{modality_name}'. "
+        f"Found layers are: {','.join(data.layers)}"
+    )
+
+# input layer argument does not work when batch_key is specified because
+# it still uses .X to filter out genes with 0 counts, even if .X might not exist.
+# So create a custom anndata as input that always uses .X
+input_layer = data.X if not par["layer"] else data.layers[par["layer"]]
+obs = pd.DataFrame(index=data.obs_names.copy())
+var = pd.DataFrame(index=data.var_names.copy())
+if par["obs_batch_key"]:
+    obs = data.obs.loc[:, par["obs_batch_key"]].to_frame()
+input_anndata = ad.AnnData(X=input_layer.copy(), obs=obs, var=var)
+if "log1p" in data.uns:
+    input_anndata.uns["log1p"] = data.uns["log1p"]
+
+# Workaround for issue
+# https://github.com/scverse/scanpy/issues/2239
+# https://github.com/scverse/scanpy/issues/2181
+if par["flavor"] != "seurat_v3":
+    # This component requires log normalized data when flavor is not seurat_v3
+    # We assume that the data is correctly normalized but scanpy will look at
+    # .uns to check the transformations performed on the data.
+    # To prevent scanpy from automatically tranforming the counts when they are
+    # already transformed, we set the appropriate values to .uns.
+    if "log1p" not in input_anndata.uns:
+        logger.warning(
+            "When flavor is not set to 'seurat_v3', "
+            "the input data for this component must be log-transformed. "
+            "However, the 'log1p' dictionairy in .uns has not been set. "
+            "This is fine if you did not log transform your data with scanpy."
+            "Otherwise, please check if you are providing log transformed "
+            "data using --layer."
+        )
+        input_anndata.uns["log1p"] = {"base": None}
+    elif "log1p" in input_anndata.uns and "base" not in input_anndata.uns["log1p"]:
+        input_anndata.uns["log1p"]["base"] = None
+
+# Enable calculating the HVG only on a subset of vars
+# e.g for cell type annotation, only calculate HVG on variables that are common between query and reference
+if par["var_input"]:
+    input_anndata.var[par["var_input"]] = data.var[par["var_input"]]
+    input_anndata = subset_vars(input_anndata, par["var_input"])
+
+logger.info("\\\\tUnfiltered data: %s", data)
+
+logger.info("\\\\tComputing hvg")
+# construct arguments
+hvg_args = {
+    "adata": input_anndata,
+    "n_top_genes": par["n_top_features"],
+    "min_mean": par["min_mean"],
+    "max_mean": par["max_mean"],
+    "min_disp": par["min_disp"],
+    "span": par["span"],
+    "n_bins": par["n_bins"],
+    "flavor": par["flavor"],
+    "subset": False,
+    "inplace": False,
+    "layer": None,  # Always uses .X because the input layer was already handled
+}
+
+optional_parameters = {
+    "max_disp": "max_disp",
+    "obs_batch_key": "batch_key",
+    "n_top_genes": "n_top_features",
+}
+# only add parameter if it's passed
+for par_name, dest_name in optional_parameters.items():
+    if par.get(par_name):
+        hvg_args[dest_name] = par[par_name]
+
+# scanpy does not do this check, although it is stated in the documentation
+if par["flavor"] == "seurat_v3" and not par["n_top_features"]:
+    raise ValueError(
+        "When flavor is set to 'seurat_v3', you are required to set 'n_top_features'."
+    )
+
+# call function
+try:
+    out = sc.pp.highly_variable_genes(**hvg_args)
+    if par["var_input"] is not None:
+        out.index = input_anndata.var.index
+        out = out.reindex(index=data.var.index, method=None)
+        out.highly_variable = out.highly_variable.fillna(False)
+        assert (out.index == data.var.index).all(), (
+            "Expected output index values to be equivalent to the input index"
+        )
+    elif par["obs_batch_key"] is not None:
+        out = out.reindex(index=data.var.index, method=None)
+        assert (out.index == data.var.index).all(), (
+            "Expected output index values to be equivalent to the input index"
+        )
+
+except ValueError as err:
+    if str(err) == "cannot specify integer \\`bins\\` when input data contains infinity":
+        err.args = (
+            "Cannot specify integer \\`bins\\` when input data contains infinity. "
+            "Perhaps input data has not been log normalized?",
+        )
+    if re.search("Bin edges must be unique:", str(err)):
+        raise RuntimeError(
+            "Scanpy failed to calculate hvg. The error "
+            "returned by scanpy (see above) could be the "
+            "result from trying to use this component on unfiltered data."
+        ) from err
+    raise err
+
+out.index = data.var.index
+logger.info("\\\\tStoring output into .var")
+if par.get("var_name_filter", None) is not None:
+    data.var[par["var_name_filter"]] = out["highly_variable"]
+
+if par.get("varm_name", None) is not None and "mean_bin" in out:
+    # drop mean_bin as mudata/anndata doesn't support tuples
+    data.varm[par["varm_name"]] = out.drop("mean_bin", axis=1)
+
+logger.info("Writing h5mu to file")
+write_h5ad_to_h5mu_with_compression(
+    par["output"], par["input"], modality_name, data, par["output_compression"]
+)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/feature_annotation/highly_variable_features_scanpy",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "singlecpu",
+    "lowmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/nextflow.config
new file mode 100644
index 0000000..5f5faeb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'feature_annotation/highly_variable_features_scanpy'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Annotate highly variable features [Satija15] [Zheng17] [Stuart19].\n\nExpects logarithmized data, except when flavor=\'seurat_v3\' in which count data is expected.\n\nDepending on flavor, this reproduces the R-implementations of Seurat [Satija15], Cell Ranger [Zheng17], and Seurat v3 [Stuart19].\n\nFor the dispersion-based methods ([Satija15] and [Zheng17]), the normalized dispersion is obtained by scaling with the mean and standard deviation of the dispersions for features falling into a given bin for mean expression of features. This means that for each bin of mean expression, highly variable features are selected.\n\nFor [Stuart19], a normalized variance for each feature is computed. First, the data are standardized (i.e., z-score normalization per feature) with a regularized standard deviation. Next, the normalized variance is computed as the variance of each feature after the transformation. Features are ranked by the normalized variance.\n'
+  author = 'Dries De Maeyer, Robrecht Cannoodt'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/nextflow_schema.json
new file mode 100644
index 0000000..83dab75
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/nextflow_schema.json
@@ -0,0 +1,142 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "highly_variable_features_scanpy",
+  "description": "Annotate highly variable features [Satija15] [Zheng17] [Stuart19].\n\nExpects logarithmized data, except when flavor='seurat_v3' in which count data is expected.\n\nDepending on flavor, this reproduces the R-implementations of Seurat [Satija15], Cell Ranger [Zheng17], and Seurat v3 [Stuart19].\n\nFor the dispersion-based methods ([Satija15] and [Zheng17]), the normalized dispersion is obtained by scaling with the mean and standard deviation of the dispersions for features falling into a given bin for mean expression of features. This means that for each bin of mean expression, highly variable features are selected.\n\nFor [Stuart19], a normalized variance for each feature is computed. First, the data are standardized (i.e., z-score normalization per feature) with a regularized standard deviation. Next, the normalized variance is computed as the variance of each feature after the transformation. Features are ranked by the normalized variance.\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "layer": {
+          "type": "string",
+          "description": "use adata.layers[layer] for expression values instead of adata.X.",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "var_input": {
+          "type": "string",
+          "description": "If specified, use boolean array in adata.var[var_input] to calculate hvg on subset of vars.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "var_name_filter": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding to which observations should be filtered out.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"filter_with_hvg\"`. ",
+          "default": "filter_with_hvg"
+        },
+        "varm_name": {
+          "type": "string",
+          "description": "In which .varm slot to store additional metadata.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"hvg\"`. ",
+          "default": "hvg"
+        },
+        "flavor": {
+          "type": "string",
+          "description": "Choose the flavor for identifying highly variable features",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"seurat\"`, choices: ``seurat`, `cell_ranger`, `seurat_v3``. ",
+          "enum": [
+            "seurat",
+            "cell_ranger",
+            "seurat_v3"
+          ],
+          "default": "seurat"
+        },
+        "n_top_features": {
+          "type": "integer",
+          "description": "Number of highly-variable features to keep",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "min_mean": {
+          "type": "number",
+          "description": "If n_top_features is defined, this and all other cutoffs for the means and the normalized dispersions are ignored",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.0125`. ",
+          "default": 0.0125
+        },
+        "max_mean": {
+          "type": "number",
+          "description": "If n_top_features is defined, this and all other cutoffs for the means and the normalized dispersions are ignored",
+          "help_text": "Type: `double`, multiple: `False`, default: `3.0`. ",
+          "default": 3.0
+        },
+        "min_disp": {
+          "type": "number",
+          "description": "If n_top_features is defined, this and all other cutoffs for the means and the normalized dispersions are ignored",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.5`. ",
+          "default": 0.5
+        },
+        "max_disp": {
+          "type": "number",
+          "description": "If n_top_features is defined, this and all other cutoffs for the means and the normalized dispersions are ignored",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "span": {
+          "type": "number",
+          "description": "The fraction of the data (cells) used when estimating the variance in the loess model fit if flavor='seurat_v3'.",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.3`. ",
+          "default": 0.3
+        },
+        "n_bins": {
+          "type": "integer",
+          "description": "Number of bins for binning the mean feature expression",
+          "help_text": "Type: `integer`, multiple: `False`, default: `20`. ",
+          "default": 20
+        },
+        "obs_batch_key": {
+          "type": "string",
+          "description": "If specified, highly-variable features are selected within each batch separately and merged",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/subset_vars.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/subset_vars.py
new file mode 100644
index 0000000..6003513
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/feature_annotation/highly_variable_features_scanpy/subset_vars.py
@@ -0,0 +1,31 @@
+def subset_vars(adata, subset_col):
+    """Subset AnnData object on highly variable genes
+
+    Parameters
+    ----------
+    adata : AnnData
+        Annotated data object
+    subset_col : str
+        Name of the boolean column in `adata.var` that contains the information if features should be used or not
+
+    Returns
+    -------
+    AnnData
+        Copy of `adata` with subsetted features
+    """
+    if subset_col not in adata.var.columns:
+        raise ValueError(
+            f"Requested to use .var column '{subset_col}' as a selection of genes, but the column is not available."
+        )
+
+    if adata.var[subset_col].dtype == "boolean":
+        assert adata.var[subset_col].isna().sum() == 0, (
+            f"The .var column `{subset_col}` contains NaN values. Can not subset data."
+        )
+        adata.var[subset_col] = adata.var[subset_col].astype("bool")
+
+    assert adata.var[subset_col].dtype == "bool", (
+        f"Expected dtype of .var column '{subset_col}' to be `bool`, but found {adata.var[subset_col].dtype}. Can not subset data."
+    )
+
+    return adata[:, adata.var[subset_col]].copy()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/.config.vsh.yaml
new file mode 100644
index 0000000..209ce8c
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/.config.vsh.yaml
@@ -0,0 +1,308 @@
+name: "delimit_fraction"
+namespace: "filter"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_fraction_column"
+    description: "Name of column from .var dataframe selecting\na column that contains\
+      \ floating point values between 0 and 1.\n"
+    info: null
+    example:
+    - "fraction_mitochondrial"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Output h5mu file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_filter"
+    description: "In which .obs slot to store a boolean array corresponding to which\
+      \ observations should be removed."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Arguments"
+  arguments:
+  - type: "double"
+    name: "--min_fraction"
+    description: "Min fraction for an observation to be retained (True in output)."
+    info: null
+    default:
+    - 0.0
+    required: false
+    min: 0.0
+    max: 1.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--max_fraction"
+    description: "Max fraction for an observation to be retained (True in output)."
+    info: null
+    default:
+    - 1.0
+    required: false
+    min: 0.0
+    max: 1.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Turns a column containing values between 0 and 1 into a boolean column\
+  \ based on thresholds.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "lowmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/filter/delimit_fraction/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/filter/delimit_fraction"
+  executable: "target/nextflow/filter/delimit_fraction/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/main.nf
new file mode 100644
index 0000000..16635d5
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/main.nf
@@ -0,0 +1,4054 @@
+// delimit_fraction v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author, maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "delimit_fraction",
+  "namespace" : "filter",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_fraction_column",
+          "description" : "Name of column from .var dataframe selecting\na column that contains floating point values between 0 and 1.\n",
+          "example" : [
+            "fraction_mitochondrial"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Output h5mu file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_filter",
+          "description" : "In which .obs slot to store a boolean array corresponding to which observations should be removed.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "double",
+          "name" : "--min_fraction",
+          "description" : "Min fraction for an observation to be retained (True in output).",
+          "default" : [
+            0.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "max" : 1.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--max_fraction",
+          "description" : "Max fraction for an observation to be retained (True in output).",
+          "default" : [
+            1.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "max" : 1.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Turns a column containing values between 0 and 1 into a boolean column based on thresholds.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "lowmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/filter/delimit_fraction/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/filter/delimit_fraction",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import mudata as mu
+import numpy as np
+import sys
+from operator import le, ge
+from pandas.api.types import is_float_dtype
+
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_fraction_column': $( if [ ! -z ${VIASH_PAR_OBS_FRACTION_COLUMN+x} ]; then echo "r'${VIASH_PAR_OBS_FRACTION_COLUMN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_name_filter': $( if [ ! -z ${VIASH_PAR_OBS_NAME_FILTER+x} ]; then echo "r'${VIASH_PAR_OBS_NAME_FILTER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'min_fraction': $( if [ ! -z ${VIASH_PAR_MIN_FRACTION+x} ]; then echo "float(r'${VIASH_PAR_MIN_FRACTION//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'max_fraction': $( if [ ! -z ${VIASH_PAR_MAX_FRACTION+x} ]; then echo "float(r'${VIASH_PAR_MAX_FRACTION//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+logger.info("Reading modality %s from %s", par["modality"], par["input"])
+data = mu.read_h5ad(par["input"], mod=par["modality"])
+
+assert data.var_names.is_unique, (
+    "The var_names of the input modality must be be unique."
+)
+
+logger.info("\\\\tUnfiltered data: %s", data)
+logger.info("\\\\tComputing aggregations.")
+
+
+def apply_filter_to_mask(mask, base, filter, comparator):
+    new_filt = np.ravel(comparator(base, filter))
+    num_removed = np.sum(np.invert(new_filt) & mask)
+    mask &= new_filt
+    return num_removed, mask
+
+
+try:
+    fraction = data.obs[par["obs_fraction_column"]]
+except KeyError:
+    raise ValueError(f"Could not find column '{par['obs_fraction_column']}'")
+if not is_float_dtype(fraction):
+    raise ValueError(
+        f"Column '{par['obs_fraction_column']}' does not contain float datatype."
+    )
+if fraction.max() > 1:
+    raise ValueError(f"Column '{par['obs_fraction_column']}' contains values > 1.")
+if fraction.min() < 0:
+    raise ValueError(f"Column '{par['obs_fraction_column']}' contains values < 0.")
+
+
+# Filter cells
+filters = (
+    (
+        "min_fraction",
+        fraction,
+        ge,
+        "\\\\tRemoving %s cells with <%s percentage mitochondrial reads.",
+    ),
+    (
+        "max_fraction",
+        fraction,
+        le,
+        "\\\\tRemoving %s cells with >%s percentage mitochondrial reads.",
+    ),
+)
+
+keep_cells = np.repeat(True, data.n_obs)
+for filter_name_or_value, base, comparator, message in filters:
+    try:
+        filter = par[filter_name_or_value]
+    except KeyError:
+        filter = filter_name_or_value
+    if filter is not None:
+        num_removed, keep_cells = apply_filter_to_mask(
+            keep_cells, base, filter, comparator
+        )
+        logger.info(message, num_removed, filter)
+
+data.obs[par["obs_name_filter"]] = keep_cells
+
+logger.info("\\\\tFiltered data: %s", data)
+logger.info("Writing output data to %s", par["output"])
+write_h5ad_to_h5mu_with_compression(
+    par["output"], par["input"], par["modality"], data, par["output_compression"]
+)
+
+logger.info("Finished")
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/filter/delimit_fraction",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "singlecpu",
+    "lowmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/nextflow.config
new file mode 100644
index 0000000..9c1d73e
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'filter/delimit_fraction'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Turns a column containing values between 0 and 1 into a boolean column based on thresholds.\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/nextflow_schema.json
new file mode 100644
index 0000000..52b00a1
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/nextflow_schema.json
@@ -0,0 +1,106 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "delimit_fraction",
+  "description": "Turns a column containing values between 0 and 1 into a boolean column based on thresholds.\n",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "obs_fraction_column": {
+          "type": "string",
+          "description": "Name of column from .var dataframe selecting\na column that contains floating point values between 0 and 1.\n",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"fraction_mitochondrial\"`. "
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "obs_name_filter": {
+          "type": "string",
+          "description": "In which .obs slot to store a boolean array corresponding to which observations should be removed.",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "min_fraction": {
+          "type": "number",
+          "description": "Min fraction for an observation to be retained (True in output).",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.0`. ",
+          "default": 0.0
+        },
+        "max_fraction": {
+          "type": "number",
+          "description": "Max fraction for an observation to be retained (True in output).",
+          "help_text": "Type: `double`, multiple: `False`, default: `1.0`. ",
+          "default": 1.0
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/delimit_fraction/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/.config.vsh.yaml
new file mode 100644
index 0000000..9d94df7
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/.config.vsh.yaml
@@ -0,0 +1,276 @@
+name: "do_filter"
+namespace: "filter"
+version: "v4.0.4"
+authors:
+- name: "Robrecht Cannoodt"
+  roles:
+  - "maintainer"
+  - "contributor"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_filter"
+    description: "Which .obs columns to use to filter the observations by."
+    info: null
+    example:
+    - "filter_with_x"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_filter"
+    description: "Which .var columns to use to filter the observations by."
+    info: null
+    example:
+    - "filter_with_x"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    description: "Output h5mu file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Remove observations and variables based on specified .obs and .var columns.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "lowmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/filter/do_filter/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/filter/do_filter"
+  executable: "target/nextflow/filter/do_filter/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/main.nf
new file mode 100644
index 0000000..a17c4eb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/main.nf
@@ -0,0 +1,3977 @@
+// do_filter v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Robrecht Cannoodt (maintainer, contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "do_filter",
+  "namespace" : "filter",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "maintainer",
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_filter",
+          "description" : "Which .obs columns to use to filter the observations by.",
+          "example" : [
+            "filter_with_x"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_filter",
+          "description" : "Which .var columns to use to filter the observations by.",
+          "example" : [
+            "filter_with_x"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Output h5mu file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Remove observations and variables based on specified .obs and .var columns.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "lowmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/filter/do_filter/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/filter/do_filter",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import mudata as mu
+import numpy as np
+import sys
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_filter': $( if [ ! -z ${VIASH_PAR_OBS_FILTER+x} ]; then echo "r'${VIASH_PAR_OBS_FILTER//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'var_filter': $( if [ ! -z ${VIASH_PAR_VAR_FILTER+x} ]; then echo "r'${VIASH_PAR_VAR_FILTER//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+logger.info("Reading %s, modality: %s", par["input"], par["modality"])
+adata = mu.read_h5ad(par["input"], mod=par["modality"])
+
+obs_filt = np.repeat(True, adata.n_obs)
+var_filt = np.repeat(True, adata.n_vars)
+
+par["obs_filter"] = par["obs_filter"] if par["obs_filter"] else []
+par["var_filter"] = par["var_filter"] if par["var_filter"] else []
+
+for obs_name in par["obs_filter"]:
+    logger.info(
+        "Filtering modality '%s' observations by .obs['%s']", par["modality"], obs_name
+    )
+    if obs_name not in adata.obs:
+        raise ValueError(f".mod[{par['modality']}].obs[{obs_name}] does not exist.")
+    if obs_name in adata.obs:
+        obs_filt &= adata.obs[obs_name]
+
+for var_name in par["var_filter"]:
+    logger.info(
+        "Filtering modality '%s' variables by .var['%s']", par["modality"], var_name
+    )
+    if var_name not in adata.var:
+        raise ValueError(f".mod[{par['modality']}].var[{var_name}] does not exist.")
+    if var_name in adata.var:
+        var_filt &= adata.var[var_name]
+
+adata_filtered = adata[obs_filt, var_filt]
+
+logger.info("Writing h5mu to file %s.", par["output"])
+write_h5ad_to_h5mu_with_compression(
+    par["output"],
+    par["input"],
+    par["modality"],
+    adata_filtered,
+    par["output_compression"],
+)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/filter/do_filter",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "singlecpu",
+    "lowmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/nextflow.config
new file mode 100644
index 0000000..bdad31a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'filter/do_filter'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Remove observations and variables based on specified .obs and .var columns.\n'
+  author = 'Robrecht Cannoodt'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/nextflow_schema.json
new file mode 100644
index 0000000..f63c9c3
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/nextflow_schema.json
@@ -0,0 +1,80 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "do_filter",
+  "description": "Remove observations and variables based on specified .obs and .var columns.\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "obs_filter": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Which .obs columns to use to filter the observations by.",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"filter_with_x\"]`. "
+        },
+        "var_filter": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Which .var columns to use to filter the observations by.",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"filter_with_x\"]`. "
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/do_filter/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/.config.vsh.yaml
new file mode 100644
index 0000000..3bfce44
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/.config.vsh.yaml
@@ -0,0 +1,369 @@
+name: "filter_with_counts"
+namespace: "filter"
+version: "v4.0.4"
+authors:
+- name: "Dries De Maeyer"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "maintainer"
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Location of the count matrix. If specified, will be used to select\
+      \ a key from .layers,\notherwise .X is used.\n"
+    info: null
+    example:
+    - "raw_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Output h5mu file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--do_subset"
+    description: "Whether to subset before storing the output."
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--obs_name_filter"
+    description: "In which .obs slot to store a boolean array corresponding to which\
+      \ observations should be removed."
+    info: null
+    default:
+    - "filter_with_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_filter"
+    description: "In which .var slot to store a boolean array corresponding to which\
+      \ variables should be removed."
+    info: null
+    default:
+    - "filter_with_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Arguments"
+  arguments:
+  - type: "integer"
+    name: "--min_counts"
+    description: "Minimum number of counts captured per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_counts"
+    description: "Maximum number of counts captured per cell."
+    info: null
+    example:
+    - 5000000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_genes_per_cell"
+    description: "Minimum of non-zero values per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_genes_per_cell"
+    description: "Maximum of non-zero values per cell."
+    info: null
+    example:
+    - 1500000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_cells_per_gene"
+    description: "Minimum of non-zero values per gene."
+    info: null
+    example:
+    - 3
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Filter scRNA-seq data based on the primary QC metrics. \nThis is based\
+  \ on both the UMI counts, the gene counts \nand the mitochondrial genes (genes starting\
+  \ with mt/MT).\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "lowmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/filter/filter_with_counts/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/filter/filter_with_counts"
+  executable: "target/nextflow/filter/filter_with_counts/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/main.nf
new file mode 100644
index 0000000..02e3c9e
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/main.nf
@@ -0,0 +1,4176 @@
+// filter_with_counts v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries De Maeyer (author)
+//  * Robrecht Cannoodt (maintainer, author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "filter_with_counts",
+  "namespace" : "filter",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "maintainer",
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Location of the count matrix. If specified, will be used to select a key from .layers,\notherwise .X is used.\n",
+          "example" : [
+            "raw_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Output h5mu file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--do_subset",
+          "description" : "Whether to subset before storing the output.",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_filter",
+          "description" : "In which .obs slot to store a boolean array corresponding to which observations should be removed.",
+          "default" : [
+            "filter_with_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_filter",
+          "description" : "In which .var slot to store a boolean array corresponding to which variables should be removed.",
+          "default" : [
+            "filter_with_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--min_counts",
+          "description" : "Minimum number of counts captured per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_counts",
+          "description" : "Maximum number of counts captured per cell.",
+          "example" : [
+            5000000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_genes_per_cell",
+          "description" : "Minimum of non-zero values per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_genes_per_cell",
+          "description" : "Maximum of non-zero values per cell.",
+          "example" : [
+            1500000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_cells_per_gene",
+          "description" : "Minimum of non-zero values per gene.",
+          "example" : [
+            3
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Filter scRNA-seq data based on the primary QC metrics. \nThis is based on both the UMI counts, the gene counts \nand the mitochondrial genes (genes starting with mt/MT).\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "lowmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/filter/filter_with_counts/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/filter/filter_with_counts",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import mudata as mu
+import numpy as np
+import sys
+from operator import le, ge, gt
+import scipy
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'layer': $( if [ ! -z ${VIASH_PAR_LAYER+x} ]; then echo "r'${VIASH_PAR_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'do_subset': $( if [ ! -z ${VIASH_PAR_DO_SUBSET+x} ]; then echo "r'${VIASH_PAR_DO_SUBSET//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'obs_name_filter': $( if [ ! -z ${VIASH_PAR_OBS_NAME_FILTER+x} ]; then echo "r'${VIASH_PAR_OBS_NAME_FILTER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_name_filter': $( if [ ! -z ${VIASH_PAR_VAR_NAME_FILTER+x} ]; then echo "r'${VIASH_PAR_VAR_NAME_FILTER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'min_counts': $( if [ ! -z ${VIASH_PAR_MIN_COUNTS+x} ]; then echo "int(r'${VIASH_PAR_MIN_COUNTS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'max_counts': $( if [ ! -z ${VIASH_PAR_MAX_COUNTS+x} ]; then echo "int(r'${VIASH_PAR_MAX_COUNTS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'min_genes_per_cell': $( if [ ! -z ${VIASH_PAR_MIN_GENES_PER_CELL+x} ]; then echo "int(r'${VIASH_PAR_MIN_GENES_PER_CELL//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'max_genes_per_cell': $( if [ ! -z ${VIASH_PAR_MAX_GENES_PER_CELL+x} ]; then echo "int(r'${VIASH_PAR_MAX_GENES_PER_CELL//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'min_cells_per_gene': $( if [ ! -z ${VIASH_PAR_MIN_CELLS_PER_GENE+x} ]; then echo "int(r'${VIASH_PAR_MIN_CELLS_PER_GENE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+logger.info("Reading input data from %s, modality %s", par["input"], par["modality"])
+modality_data = mu.read_h5ad(par["input"], mod=par["modality"])
+
+logger.info("\\\\tUnfiltered data: %s", modality_data)
+
+layer_name = "X" if not par["layer"] else par["layer"]
+logger.info("Selecting input layer %s", layer_name)
+input_layer = (
+    modality_data.X if not par["layer"] else modality_data.layers[par["layer"]]
+)
+if scipy.sparse.issparse(input_layer):
+    # Below is a check that scipy does not do with check_format; see https://github.com/scipy/scipy/issues/23784
+    # also, check_format might adjust the matrix attributes in place; so do this check before check_format
+
+    assert input_layer.nnz == input_layer.data.size, (
+        f"The provided sparse matrix (i.e. {layer_name} from modality {par['modality']}) is malformatted. "
+        f"The number of elements that the matrix should hold (i.e. .nnz={input_layer.nnz}) does not correspond with "
+        f"the number of elements actually being stored (i.e. .data.size={input_layer.data.size})"
+    )
+    input_layer.check_format(full_check=True)
+
+
+logger.info("\\\\tComputing aggregations.")
+n_counts_per_cell = np.ravel(np.sum(input_layer, axis=1))
+n_cells_per_gene = np.sum(input_layer > 0, axis=0)
+n_genes_per_cell = np.sum(input_layer > 0, axis=1)
+
+
+def apply_filter_to_mask(mask, base, filter, comparator):
+    new_filt = np.ravel(comparator(base, filter))
+    num_removed = np.sum(np.invert(new_filt) & mask)
+    mask &= new_filt
+    return num_removed, mask
+
+
+# Filter genes
+keep_genes = np.repeat(True, modality_data.n_vars)
+if par["min_cells_per_gene"] is not None:
+    num_removed, keep_genes = apply_filter_to_mask(
+        keep_genes, n_cells_per_gene, par["min_cells_per_gene"], ge
+    )
+    logger.info(
+        "\\\\tRemoving %s genes with non-zero values in <%s cells.",
+        num_removed,
+        par["min_cells_per_gene"],
+    )
+
+# Filter cells
+filters = (
+    (
+        "min_genes_per_cell",
+        n_genes_per_cell,
+        ge,
+        "\\\\tRemoving %s cells with non-zero values in <%s genes.",
+    ),
+    (
+        "max_genes_per_cell",
+        n_genes_per_cell,
+        le,
+        "\\\\tRemoving %s cells with non-zero values in >%s genes.",
+    ),
+    ("min_counts", n_counts_per_cell, ge, "\\\\tRemoving %s cells with <%s total counts."),
+    ("max_counts", n_counts_per_cell, le, "\\\\tRemoving %s cells with >%s total counts."),
+    (
+        0,
+        np.sum(input_layer[:, keep_genes], axis=1),
+        gt,
+        "\\\\tRemoving %s cells with %s counts",
+    ),
+)
+
+keep_cells = np.repeat(True, modality_data.n_obs)
+for filter_name_or_value, base, comparator, message in filters:
+    try:
+        filter = par[filter_name_or_value]
+    except KeyError:
+        filter = filter_name_or_value
+    if filter is not None:
+        num_removed, keep_cells = apply_filter_to_mask(
+            keep_cells, base, filter, comparator
+        )
+        logger.info(message, num_removed, filter)
+
+if par["obs_name_filter"] is not None:
+    modality_data.obs[par["obs_name_filter"]] = keep_cells
+if par["var_name_filter"] is not None:
+    modality_data.var[par["var_name_filter"]] = keep_genes
+
+if par["do_subset"]:
+    modality_data = modality_data[keep_cells, keep_genes]
+
+logger.info("\\\\tFiltered data: %s", modality_data)
+logger.info(
+    "Writing output data to %s with compression %s",
+    par["output"],
+    par["output_compression"],
+)
+write_h5ad_to_h5mu_with_compression(
+    par["output"],
+    par["input"],
+    par["modality"],
+    modality_data,
+    par["output_compression"],
+)
+
+
+logger.info("Finished")
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/filter/filter_with_counts",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "singlecpu",
+    "lowmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/nextflow.config
new file mode 100644
index 0000000..be049ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'filter/filter_with_counts'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Filter scRNA-seq data based on the primary QC metrics. \nThis is based on both the UMI counts, the gene counts \nand the mitochondrial genes (genes starting with mt/MT).\n'
+  author = 'Dries De Maeyer, Robrecht Cannoodt'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/nextflow_schema.json
new file mode 100644
index 0000000..4e586c5
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/nextflow_schema.json
@@ -0,0 +1,132 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "filter_with_counts",
+  "description": "Filter scRNA-seq data based on the primary QC metrics. \nThis is based on both the UMI counts, the gene counts \nand the mitochondrial genes (genes starting with mt/MT).\n",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "layer": {
+          "type": "string",
+          "description": "Location of the count matrix",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"raw_counts\"`. "
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "do_subset": {
+          "type": "boolean",
+          "description": "Whether to subset before storing the output.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "obs_name_filter": {
+          "type": "string",
+          "description": "In which .obs slot to store a boolean array corresponding to which observations should be removed.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"filter_with_counts\"`. ",
+          "default": "filter_with_counts"
+        },
+        "var_name_filter": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding to which variables should be removed.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"filter_with_counts\"`. ",
+          "default": "filter_with_counts"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "min_counts": {
+          "type": "integer",
+          "description": "Minimum number of counts captured per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "max_counts": {
+          "type": "integer",
+          "description": "Maximum number of counts captured per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `5000000`. "
+        },
+        "min_genes_per_cell": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "max_genes_per_cell": {
+          "type": "integer",
+          "description": "Maximum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `1500000`. "
+        },
+        "min_cells_per_gene": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per gene.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3`. "
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_counts/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/.config.vsh.yaml
new file mode 100644
index 0000000..ddfdf2c
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/.config.vsh.yaml
@@ -0,0 +1,426 @@
+name: "filter_with_scrublet"
+namespace: "filter"
+version: "v4.0.4"
+authors:
+- name: "Dries De Maeyer"
+  roles:
+  - "contributor"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "maintainer"
+  - "contributor"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Input layer to use as data for calculating doublets. .X is used\
+      \ not specified."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    description: "Output h5mu file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_filter"
+    description: "In which .obs slot to store a boolean array corresponding to which\
+      \ observations should be filtered out."
+    info: null
+    default:
+    - "filter_with_scrublet"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--do_subset"
+    description: "Whether to subset before storing the output."
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--obs_name_doublet_score"
+    description: "Name of the doublet scores column in the obs slot of the returned\
+      \ object."
+    info: null
+    default:
+    - "scrublet_doublet_score"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--expected_doublet_rate"
+    description: "The estimated fraction of doublets as from the experimental setup.\n"
+    info: null
+    required: false
+    min: 0.0
+    max: 1.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--stdev_doublet_rate"
+    description: "Uncertainty in the expected doublet rate."
+    info: null
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--n_neighbors"
+    description: "Number of neighbors used to construct the KNN classifier of observed\
+      \ transcriptomes\nand simulated doublets.\n"
+    info: null
+    required: false
+    min: 0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--sim_doublet_ratio"
+    description: "Number of doublets to simulate relative to the number of observed\n\
+      transcriptomes.\n"
+    info: null
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_counts"
+    description: "The number of minimal UMI counts per cell that have to be present\
+      \ for initial cell detection."
+    info: null
+    default:
+    - 2
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_cells"
+    description: "The number of cells in which UMIs for a gene were detected."
+    info: null
+    default:
+    - 3
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--min_gene_variablity_percent"
+    description: "Used for gene filtering prior to PCA. Keep the most highly variable\
+      \ genes (in the top min_gene_variability_pctl percentile), as measured by the\
+      \ v-statistic [Klein et al., Cell 2015]."
+    info: null
+    default:
+    - 85.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--num_pca_components"
+    description: "Number of principal components to use during PCA dimensionality\
+      \ reduction."
+    info: null
+    default:
+    - 30
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--distance_metric"
+    description: "The distance metric used for computing similarities."
+    info: null
+    default:
+    - "euclidean"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--allow_automatic_threshold_detection_fail"
+    description: "When scrublet fails to automatically determine the double score\
+      \ threshold, \nallow the component to continue and set the output columns to\
+      \ NA.\n"
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Doublet detection using the Scrublet method (Wolock, Lopez and Klein,\
+  \ 2019).\nThe method tests for potential doublets by using the expression profiles\
+  \ of\ncells to generate synthetic potential doubles which are tested against cells.\
+  \ \nThe method returns a \"doublet score\" on which it calls for potential doublets.\n\
+  \nFor the source code please visit https://github.com/AllonKleinLab/scrublet.\n\n\
+  For 10x we expect the doublet rates to be:\n  Multiplet Rate (%) - # of Cells Loaded\
+  \ - # of Cells Recovered\n  ~0.4% ~800 ~500\n  ~0.8% ~1,600 ~1,000\n  ~1.6% ~3,200\
+  \ ~2,000\n  ~2.3% ~4,800 ~3,000\n  ~3.1% ~6,400 ~4,000\n  ~3.9% ~8,000 ~5,000\n\
+  \  ~4.6% ~9,600 ~6,000\n  ~5.4% ~11,200 ~7,000\n  ~6.1% ~12,800 ~8,000\n  ~6.9%\
+  \ ~14,400 ~9,000\n  ~7.6% ~16,000 ~10,000\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+  docker_run_args:
+  - "--env NUMBA_CACHE_DIR=/tmp"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "highcpu"
+    - "midmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.13-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    - "build-essential"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    - "scrublet"
+    - "annoy==1.17.3"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/filter/filter_with_scrublet/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/filter/filter_with_scrublet"
+  executable: "target/nextflow/filter/filter_with_scrublet/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/main.nf
new file mode 100644
index 0000000..7d45ddf
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/main.nf
@@ -0,0 +1,4191 @@
+// filter_with_scrublet v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries De Maeyer (contributor)
+//  * Robrecht Cannoodt (maintainer, contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "filter_with_scrublet",
+  "namespace" : "filter",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "maintainer",
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Input layer to use as data for calculating doublets. .X is used not specified.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Output h5mu file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_filter",
+          "description" : "In which .obs slot to store a boolean array corresponding to which observations should be filtered out.",
+          "default" : [
+            "filter_with_scrublet"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--do_subset",
+          "description" : "Whether to subset before storing the output.",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_doublet_score",
+          "description" : "Name of the doublet scores column in the obs slot of the returned object.",
+          "default" : [
+            "scrublet_doublet_score"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--expected_doublet_rate",
+          "description" : "The estimated fraction of doublets as from the experimental setup.\n",
+          "required" : false,
+          "min" : 0.0,
+          "max" : 1.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--stdev_doublet_rate",
+          "description" : "Uncertainty in the expected doublet rate.",
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--n_neighbors",
+          "description" : "Number of neighbors used to construct the KNN classifier of observed transcriptomes\nand simulated doublets.\n",
+          "required" : false,
+          "min" : 0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--sim_doublet_ratio",
+          "description" : "Number of doublets to simulate relative to the number of observed\ntranscriptomes.\n",
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_counts",
+          "description" : "The number of minimal UMI counts per cell that have to be present for initial cell detection.",
+          "default" : [
+            2
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_cells",
+          "description" : "The number of cells in which UMIs for a gene were detected.",
+          "default" : [
+            3
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--min_gene_variablity_percent",
+          "description" : "Used for gene filtering prior to PCA. Keep the most highly variable genes (in the top min_gene_variability_pctl percentile), as measured by the v-statistic [Klein et al., Cell 2015].",
+          "default" : [
+            85.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--num_pca_components",
+          "description" : "Number of principal components to use during PCA dimensionality reduction.",
+          "default" : [
+            30
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--distance_metric",
+          "description" : "The distance metric used for computing similarities.",
+          "default" : [
+            "euclidean"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--allow_automatic_threshold_detection_fail",
+          "description" : "When scrublet fails to automatically determine the double score threshold, \nallow the component to continue and set the output columns to NA.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Doublet detection using the Scrublet method (Wolock, Lopez and Klein, 2019).\nThe method tests for potential doublets by using the expression profiles of\ncells to generate synthetic potential doubles which are tested against cells. \nThe method returns a \\"doublet score\\" on which it calls for potential doublets.\n\nFor the source code please visit https://github.com/AllonKleinLab/scrublet.\n\nFor 10x we expect the doublet rates to be:\n  Multiplet Rate (%) - # of Cells Loaded - # of Cells Recovered\n  ~0.4% ~800 ~500\n  ~0.8% ~1,600 ~1,000\n  ~1.6% ~3,200 ~2,000\n  ~2.3% ~4,800 ~3,000\n  ~3.1% ~6,400 ~4,000\n  ~3.9% ~8,000 ~5,000\n  ~4.6% ~9,600 ~6,000\n  ~5.4% ~11,200 ~7,000\n  ~6.1% ~12,800 ~8,000\n  ~6.9% ~14,400 ~9,000\n  ~7.6% ~16,000 ~10,000\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild",
+      "docker_run_args" : [
+        "--env NUMBA_CACHE_DIR=/tmp"
+      ]
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "highcpu",
+          "midmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.13-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps",
+            "build-essential"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4",
+            "scrublet",
+            "annoy==1.17.3"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/filter/filter_with_scrublet/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/filter/filter_with_scrublet",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import scrublet as scr
+import mudata as mu
+import numpy as np
+import sys
+import pandas as pd
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'layer': $( if [ ! -z ${VIASH_PAR_LAYER+x} ]; then echo "r'${VIASH_PAR_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_name_filter': $( if [ ! -z ${VIASH_PAR_OBS_NAME_FILTER+x} ]; then echo "r'${VIASH_PAR_OBS_NAME_FILTER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'do_subset': $( if [ ! -z ${VIASH_PAR_DO_SUBSET+x} ]; then echo "r'${VIASH_PAR_DO_SUBSET//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'obs_name_doublet_score': $( if [ ! -z ${VIASH_PAR_OBS_NAME_DOUBLET_SCORE+x} ]; then echo "r'${VIASH_PAR_OBS_NAME_DOUBLET_SCORE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'expected_doublet_rate': $( if [ ! -z ${VIASH_PAR_EXPECTED_DOUBLET_RATE+x} ]; then echo "float(r'${VIASH_PAR_EXPECTED_DOUBLET_RATE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'stdev_doublet_rate': $( if [ ! -z ${VIASH_PAR_STDEV_DOUBLET_RATE+x} ]; then echo "float(r'${VIASH_PAR_STDEV_DOUBLET_RATE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'n_neighbors': $( if [ ! -z ${VIASH_PAR_N_NEIGHBORS+x} ]; then echo "int(r'${VIASH_PAR_N_NEIGHBORS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'sim_doublet_ratio': $( if [ ! -z ${VIASH_PAR_SIM_DOUBLET_RATIO+x} ]; then echo "float(r'${VIASH_PAR_SIM_DOUBLET_RATIO//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'min_counts': $( if [ ! -z ${VIASH_PAR_MIN_COUNTS+x} ]; then echo "int(r'${VIASH_PAR_MIN_COUNTS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'min_cells': $( if [ ! -z ${VIASH_PAR_MIN_CELLS+x} ]; then echo "int(r'${VIASH_PAR_MIN_CELLS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'min_gene_variablity_percent': $( if [ ! -z ${VIASH_PAR_MIN_GENE_VARIABLITY_PERCENT+x} ]; then echo "float(r'${VIASH_PAR_MIN_GENE_VARIABLITY_PERCENT//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'num_pca_components': $( if [ ! -z ${VIASH_PAR_NUM_PCA_COMPONENTS+x} ]; then echo "int(r'${VIASH_PAR_NUM_PCA_COMPONENTS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'distance_metric': $( if [ ! -z ${VIASH_PAR_DISTANCE_METRIC+x} ]; then echo "r'${VIASH_PAR_DISTANCE_METRIC//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'allow_automatic_threshold_detection_fail': $( if [ ! -z ${VIASH_PAR_ALLOW_AUTOMATIC_THRESHOLD_DETECTION_FAIL+x} ]; then echo "r'${VIASH_PAR_ALLOW_AUTOMATIC_THRESHOLD_DETECTION_FAIL//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+
+logger = setup_logger()
+
+logger.info("Reading %s, modality %s", par["input"], par["modality"])
+data = mu.read_h5ad(par["input"], mod=par["modality"])
+
+logger.info("Using layer '%s'.", "X" if not par["layer"] else par["layer"])
+input_layer = data.X if not par["layer"] else data.layers[par["layer"]]
+
+if 0 in input_layer.shape:
+    raise ValueError(
+        f"Modality {par['modality']} of input Mudata {par['input']} appears "
+        f"to be empty (shape: {input_layer.shape})."
+    )
+
+logger.info("\\\\tRunning scrublet")
+initializer_args = {
+    arg: par[arg]
+    for arg in (
+        "expected_doublet_rate",
+        "stdev_doublet_rate",
+        "n_neighbors",
+        "sim_doublet_ratio",
+    )
+    if par[arg] is not None
+}
+scrub = scr.Scrublet(input_layer, **initializer_args)
+
+doublet_scores, predicted_doublets = scrub.scrub_doublets(
+    min_counts=par["min_counts"],
+    min_cells=par["min_cells"],
+    min_gene_variability_pctl=par["min_gene_variablity_percent"],
+    n_prin_comps=par["num_pca_components"],
+    distance_metric=par["distance_metric"],
+    use_approx_neighbors=False,
+)
+
+try:
+    keep_cells = np.invert(predicted_doublets)
+except TypeError:
+    if par["allow_automatic_threshold_detection_fail"]:
+        # Scrublet might not throw an error and return None if it fails to detect doublets...
+        logger.info(
+            "\\\\tScrublet could not automatically detect the doublet score threshold. Setting output columns to NA."
+        )
+        keep_cells = np.nan
+        doublet_scores = np.nan
+    else:
+        raise RuntimeError(
+            "Scrublet could not automatically detect the doublet score threshold. "
+            "--allow_automatic_threshold_detection_fail can be used to ignore this failure "
+            "and set the corresponding output columns to NA."
+        )
+
+logger.info("\\\\tStoring output into .obs")
+if par["obs_name_doublet_score"] is not None:
+    data.obs[par["obs_name_doublet_score"]] = doublet_scores
+    data.obs[par["obs_name_doublet_score"]] = data.obs[
+        par["obs_name_doublet_score"]
+    ].astype("float64")
+if par["obs_name_filter"] is not None:
+    data.obs[par["obs_name_filter"]] = keep_cells
+    data.obs[par["obs_name_filter"]] = data.obs[par["obs_name_filter"]].astype(
+        pd.BooleanDtype()
+    )
+
+if par["do_subset"]:
+    if pd.api.types.is_scalar(keep_cells) and pd.isna(keep_cells):
+        logger.warning("Not subsetting beacuse doublets were not predicted")
+    else:
+        data = data[keep_cells, :]
+
+logger.info(
+    "Writing h5mu to %s, with compression %s", par["output"], par["output_compression"]
+)
+write_h5ad_to_h5mu_with_compression(
+    par["output"], par["input"], par["modality"], data, par["output_compression"]
+)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/filter/filter_with_scrublet",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "highcpu",
+    "midmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/nextflow.config
new file mode 100644
index 0000000..931f170
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'filter/filter_with_scrublet'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Doublet detection using the Scrublet method (Wolock, Lopez and Klein, 2019).\nThe method tests for potential doublets by using the expression profiles of\ncells to generate synthetic potential doubles which are tested against cells. \nThe method returns a "doublet score" on which it calls for potential doublets.\n\nFor the source code please visit https://github.com/AllonKleinLab/scrublet.\n\nFor 10x we expect the doublet rates to be:\n  Multiplet Rate (%) - # of Cells Loaded - # of Cells Recovered\n  ~0.4% ~800 ~500\n  ~0.8% ~1,600 ~1,000\n  ~1.6% ~3,200 ~2,000\n  ~2.3% ~4,800 ~3,000\n  ~3.1% ~6,400 ~4,000\n  ~3.9% ~8,000 ~5,000\n  ~4.6% ~9,600 ~6,000\n  ~5.4% ~11,200 ~7,000\n  ~6.1% ~12,800 ~8,000\n  ~6.9% ~14,400 ~9,000\n  ~7.6% ~16,000 ~10,000\n'
+  author = 'Dries De Maeyer, Robrecht Cannoodt'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/nextflow_schema.json
new file mode 100644
index 0000000..dae9f36
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/nextflow_schema.json
@@ -0,0 +1,143 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "filter_with_scrublet",
+  "description": "Doublet detection using the Scrublet method (Wolock, Lopez and Klein, 2019).\nThe method tests for potential doublets by using the expression profiles of\ncells to generate synthetic potential doubles which are tested against cells. \nThe method returns a \"doublet score\" on which it calls for potential doublets.\n\nFor the source code please visit https://github.com/AllonKleinLab/scrublet.\n\nFor 10x we expect the doublet rates to be:\n  Multiplet Rate (%) - # of Cells Loaded - # of Cells Recovered\n  ~0.4% ~800 ~500\n  ~0.8% ~1,600 ~1,000\n  ~1.6% ~3,200 ~2,000\n  ~2.3% ~4,800 ~3,000\n  ~3.1% ~6,400 ~4,000\n  ~3.9% ~8,000 ~5,000\n  ~4.6% ~9,600 ~6,000\n  ~5.4% ~11,200 ~7,000\n  ~6.1% ~12,800 ~8,000\n  ~6.9% ~14,400 ~9,000\n  ~7.6% ~16,000 ~10,000\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "layer": {
+          "type": "string",
+          "description": "Input layer to use as data for calculating doublets",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "obs_name_filter": {
+          "type": "string",
+          "description": "In which .obs slot to store a boolean array corresponding to which observations should be filtered out.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"filter_with_scrublet\"`. ",
+          "default": "filter_with_scrublet"
+        },
+        "do_subset": {
+          "type": "boolean",
+          "description": "Whether to subset before storing the output.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "obs_name_doublet_score": {
+          "type": "string",
+          "description": "Name of the doublet scores column in the obs slot of the returned object.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scrublet_doublet_score\"`. ",
+          "default": "scrublet_doublet_score"
+        },
+        "expected_doublet_rate": {
+          "type": "number",
+          "description": "The estimated fraction of doublets as from the experimental setup.\n",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "stdev_doublet_rate": {
+          "type": "number",
+          "description": "Uncertainty in the expected doublet rate.",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "n_neighbors": {
+          "type": "integer",
+          "description": "Number of neighbors used to construct the KNN classifier of observed transcriptomes\nand simulated doublets.\n",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "sim_doublet_ratio": {
+          "type": "number",
+          "description": "Number of doublets to simulate relative to the number of observed\ntranscriptomes.\n",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "min_counts": {
+          "type": "integer",
+          "description": "The number of minimal UMI counts per cell that have to be present for initial cell detection.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `2`. ",
+          "default": 2
+        },
+        "min_cells": {
+          "type": "integer",
+          "description": "The number of cells in which UMIs for a gene were detected.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `3`. ",
+          "default": 3
+        },
+        "min_gene_variablity_percent": {
+          "type": "number",
+          "description": "Used for gene filtering prior to PCA",
+          "help_text": "Type: `double`, multiple: `False`, default: `85.0`. ",
+          "default": 85.0
+        },
+        "num_pca_components": {
+          "type": "integer",
+          "description": "Number of principal components to use during PCA dimensionality reduction.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `30`. ",
+          "default": 30
+        },
+        "distance_metric": {
+          "type": "string",
+          "description": "The distance metric used for computing similarities.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"euclidean\"`. ",
+          "default": "euclidean"
+        },
+        "allow_automatic_threshold_detection_fail": {
+          "type": "boolean",
+          "description": "When scrublet fails to automatically determine the double score threshold, \nallow the component to continue and set the output columns to NA.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/filter/filter_with_scrublet/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/.config.vsh.yaml
new file mode 100644
index 0000000..e23f14a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/.config.vsh.yaml
@@ -0,0 +1,323 @@
+name: "harmonypy"
+namespace: "integrate"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "contributor"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_input"
+    description: "Which .obsm slot to use as a starting PCA embedding."
+    info: null
+    default:
+    - "X_pca"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_output"
+    description: "In which .obsm slot to store the resulting integrated embedding."
+    info: null
+    default:
+    - "X_pca_integrated"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--theta"
+    description: "Diversity clustering penalty parameter. Can be set as a single value\
+      \ for all batch observations or as multiple values, one for each observation\
+      \ in the batches defined by --obs_covariates. theta=0 does not encourage any\
+      \ diversity. Larger values of theta result in more diverse clusters."
+    info: null
+    default:
+    - 2.0
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_covariates"
+    description: "The .obs field(s) that define the covariate(s) to regress out."
+    info: null
+    example:
+    - "batch"
+    - "sample"
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Performs Harmony integration based as described in https://github.com/immunogenomics/harmony.\
+  \ Based on an implementation in python from https://github.com/slowkow/harmonypy"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3_mms.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "highmem"
+    - "highcpu"
+    - "highdisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    - "harmonypy~=0.0.6"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/integrate/harmonypy/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/integrate/harmonypy"
+  executable: "target/nextflow/integrate/harmonypy/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/main.nf
new file mode 100644
index 0000000..1c8f9ac
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/main.nf
@@ -0,0 +1,4031 @@
+// harmonypy v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+//  * Robrecht Cannoodt (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "harmonypy",
+  "namespace" : "integrate",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_input",
+          "description" : "Which .obsm slot to use as a starting PCA embedding.",
+          "default" : [
+            "X_pca"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_output",
+          "description" : "In which .obsm slot to store the resulting integrated embedding.",
+          "default" : [
+            "X_pca_integrated"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--theta",
+          "description" : "Diversity clustering penalty parameter. Can be set as a single value for all batch observations or as multiple values, one for each observation in the batches defined by --obs_covariates. theta=0 does not encourage any diversity. Larger values of theta result in more diverse clusters.",
+          "default" : [
+            2.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_covariates",
+          "description" : "The .obs field(s) that define the covariate(s) to regress out.",
+          "example" : [
+            "batch",
+            "sample"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Performs Harmony integration based as described in https://github.com/immunogenomics/harmony. Based on an implementation in python from https://github.com/slowkow/harmonypy",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_mms.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "highmem",
+          "highcpu",
+          "highdisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4",
+            "harmonypy~=0.0.6"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/integrate/harmonypy/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/integrate/harmonypy",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import mudata
+import sys
+from harmonypy import run_harmony
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_input': $( if [ ! -z ${VIASH_PAR_OBSM_INPUT+x} ]; then echo "r'${VIASH_PAR_OBSM_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_output': $( if [ ! -z ${VIASH_PAR_OBSM_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OBSM_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'theta': $( if [ ! -z ${VIASH_PAR_THETA+x} ]; then echo "list(map(float, r'${VIASH_PAR_THETA//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'obs_covariates': $( if [ ! -z ${VIASH_PAR_OBS_COVARIATES+x} ]; then echo "r'${VIASH_PAR_OBS_COVARIATES//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+
+def main():
+    logger.info("Reading %s from %s", par["modality"], par["input"])
+    mod = mudata.read_h5ad(par["input"].strip(), mod=par["modality"])
+    logger.info("Selecting embedding %s", par["obsm_input"])
+    pca_embedding = mod.obsm[par["obsm_input"]]
+    metadata = mod.obs
+    logger.info(
+        "Available covariates to regress out: ", ", ".join(metadata.columns.to_list())
+    )
+    logger.info(
+        "Running harmonypy with theta %s to regress out the following variables: ",
+        par["theta"],
+        ", ".join(par["obs_covariates"]),
+    )
+    ho = run_harmony(pca_embedding, metadata, par["obs_covariates"], theta=par["theta"])
+    logger.info("Adding output to slot %s", par["obsm_output"])
+    mod.obsm[par["obsm_output"]] = ho.Z_corr.T
+    logger.info("Writing to %s")
+    write_h5ad_to_h5mu_with_compression(
+        par["output"], par["input"], par["modality"], mod, par["output_compression"]
+    )
+    logger.info("Finished")
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/integrate/harmonypy",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "highmem",
+    "highcpu",
+    "highdisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/nextflow.config
new file mode 100644
index 0000000..0b995f9
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'integrate/harmonypy'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Performs Harmony integration based as described in https://github.com/immunogenomics/harmony. Based on an implementation in python from https://github.com/slowkow/harmonypy'
+  author = 'Dries Schaumont, Robrecht Cannoodt'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/nextflow_schema.json
new file mode 100644
index 0000000..20d00b7
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/nextflow_schema.json
@@ -0,0 +1,95 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "harmonypy",
+  "description": "Performs Harmony integration based as described in https://github.com/immunogenomics/harmony. Based on an implementation in python from https://github.com/slowkow/harmonypy",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`. "
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output\"`, direction: `output`. ",
+          "default": "$id.$key.output"
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "obsm_input": {
+          "type": "string",
+          "description": "Which .obsm slot to use as a starting PCA embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_pca\"`. ",
+          "default": "X_pca"
+        },
+        "obsm_output": {
+          "type": "string",
+          "description": "In which .obsm slot to store the resulting integrated embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_pca_integrated\"`. ",
+          "default": "X_pca_integrated"
+        },
+        "theta": {
+          "type": "array",
+          "items": {
+            "type": "number"
+          },
+          "description": "Diversity clustering penalty parameter",
+          "help_text": "Type: `double`, multiple: `True`, default: `[2.0]`. ",
+          "default": [
+            2.0
+          ]
+        },
+        "obs_covariates": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The .obs field(s) that define the covariate(s) to regress out.",
+          "help_text": "Type: `string`, multiple: `True`, required, example: `[\"batch\";\"sample\"]`. "
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/harmonypy/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/.config.vsh.yaml
new file mode 100644
index 0000000..3fe787e
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/.config.vsh.yaml
@@ -0,0 +1,522 @@
+name: "scarches"
+namespace: "integrate"
+version: "v4.0.4"
+authors:
+- name: "Vladimir Shitov"
+  roles:
+  - "author"
+  info:
+    role: "Contributor"
+    links:
+      email: "vladimir.shitov@helmholtz-muenchen.de"
+      github: "vladimirshitov"
+      orcid: "0000-0002-1960-8812"
+      linkedin: "vladimir-shitov-9a659513b"
+    organizations:
+    - name: "Helmholtz Munich"
+      href: "https://www.helmholtz-munich.de"
+      role: "PhD Candidate"
+- name: "Dorien Roosen"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  description: "Arguments related to the input (query) dataset"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file to use as a query"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Layer to be used for scArches, if .X is not to be used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_obs_batch"
+    description: "Name of the .obs column with batch information."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_obs_label"
+    description: "Name of the .obs column with celltype information."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_var_gene_names"
+    description: "Name of the .var column with gene names, if the var .index is not\
+      \ to be used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_obs_size_factor"
+    description: "Key in adata.obs for size factor information. Instead of using library\
+      \ size as a size factor,\nthe provided size factor column will be used as offset\
+      \ in the mean of the likelihood.\nAssumed to be on linear scale.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_obs_categorical_covariate"
+    description: "Keys in adata.obs that correspond to categorical data. These covariates\
+      \ can be added in\naddition to the batch covariate and are also treated as nuisance\
+      \ factors\n(i.e., the model tries to minimize their effects on the latent space).\n\
+      Thus, these should not be used for biologically-relevant factors that you do\
+      \ _not_ want to correct for.\nImportant: the order of the categorical covariates\
+      \ matters and should match the order of the covariates in the trained reference\
+      \ model.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_obs_continuous_covariate"
+    description: "Keys in adata.obs that correspond to continuous data. These covariates\
+      \ can be added in\naddition to the batch covariate and are also treated as nuisance\
+      \ factors\n(i.e., the model tries to minimize their effects on the latent space).\
+      \ Thus, these should not be\nused for biologically-relevant factors that you\
+      \ do _not_ want to correct for.\nImportant: the order of the continuous covariates\
+      \ matters and should match the order of the covariates in the trained reference\
+      \ model.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--sanitize_ensembl_ids"
+    description: "Whether to sanitize ensembl ids by removing version numbers."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Reference"
+  arguments:
+  - type: "file"
+    name: "--reference"
+    alternatives:
+    - "-r"
+    description: "Path to the directory with reference model or a web link."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_class"
+    description: "For legacy models; the type of model (where the type of model was\
+      \ not saved with it; e.g. when they were generated with scvi-tools versions\
+      \ < 0.15)."
+    info: null
+    example:
+    - "SCANVI"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--model_output"
+    description: "Output directory for model"
+    info: null
+    default:
+    - "model"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_output"
+    description: "In which .obsm slot to store the resulting integrated embedding."
+    info: null
+    default:
+    - "X_integrated_scanvi"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_output_predictions"
+    description: "In which .obs slot to store the resulting label predictions. Only\
+      \ relevant if a scANVI model was provided."
+    info: null
+    default:
+    - "scanvi_pred"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_output_probabilities"
+    description: "In which .obs slot to store the probabilities of the label predictions.\
+      \ Only relevant if a scANVI model was provided."
+    info: null
+    default:
+    - "scanvi_proba"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Early stopping arguments"
+  arguments:
+  - type: "boolean"
+    name: "--early_stopping"
+    description: "Whether to perform early stopping with respect to the validation\
+      \ set."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--early_stopping_monitor"
+    description: "Metric logged during validation set epoch."
+    info: null
+    default:
+    - "elbo_validation"
+    required: false
+    choices:
+    - "elbo_validation"
+    - "reconstruction_loss_validation"
+    - "kl_local_validation"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--early_stopping_patience"
+    description: "Number of validation epochs with no improvement after which training\
+      \ will be stopped."
+    info: null
+    default:
+    - 45
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--early_stopping_min_delta"
+    description: "Minimum change in the monitored quantity to qualify as an improvement,\
+      \ i.e. an absolute change of less than min_delta, will count as no improvement."
+    info: null
+    default:
+    - 0.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Learning parameters"
+  arguments:
+  - type: "integer"
+    name: "--max_epochs"
+    description: "Number of passes through the dataset, defaults to (20000 / number\
+      \ of cells) * 400 or 400; whichever is smallest."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--reduce_lr_on_plateau"
+    description: "Whether to monitor validation loss and reduce learning rate when\
+      \ validation set `lr_scheduler_metric` plateaus."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--lr_factor"
+    description: "Factor to reduce learning rate."
+    info: null
+    default:
+    - 0.6
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--lr_patience"
+    description: "Number of epochs with no improvement after which learning rate will\
+      \ be reduced."
+    info: null
+    default:
+    - 30.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "set_var_index.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Performs reference mapping with scArches"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "scanvi_model"
+- type: "file"
+  path: "scanvi_covariate_model"
+- type: "file"
+  path: "scvi_model"
+- type: "file"
+  path: "pbmc_1k_protein_v3_mms.h5mu"
+- type: "file"
+  path: "HLCA_reference_model.zip"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "highmem"
+    - "highcpu"
+    - "highdisk"
+    - "gpu"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "pytorch/pytorch:2.9.1-cuda12.8-cudnn9-runtime"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scvi-tools[cuda]~=1.4.1"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/integrate/scarches/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/integrate/scarches"
+  executable: "target/nextflow/integrate/scarches/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/main.nf
new file mode 100644
index 0000000..2a424f9
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/main.nf
@@ -0,0 +1,4593 @@
+// scarches v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Vladimir Shitov (author)
+//  * Dorien Roosen (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "scarches",
+  "namespace" : "integrate",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Vladimir Shitov",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "links" : {
+          "email" : "vladimir.shitov@helmholtz-muenchen.de",
+          "github" : "vladimirshitov",
+          "orcid" : "0000-0002-1960-8812",
+          "linkedin" : "vladimir-shitov-9a659513b"
+        },
+        "organizations" : [
+          {
+            "name" : "Helmholtz Munich",
+            "href" : "https://www.helmholtz-munich.de",
+            "role" : "PhD Candidate"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "description" : "Arguments related to the input (query) dataset",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file to use as a query",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Layer to be used for scArches, if .X is not to be used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_obs_batch",
+          "description" : "Name of the .obs column with batch information.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_obs_label",
+          "description" : "Name of the .obs column with celltype information.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_var_gene_names",
+          "description" : "Name of the .var column with gene names, if the var .index is not to be used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_obs_size_factor",
+          "description" : "Key in adata.obs for size factor information. Instead of using library size as a size factor,\nthe provided size factor column will be used as offset in the mean of the likelihood.\nAssumed to be on linear scale.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_obs_categorical_covariate",
+          "description" : "Keys in adata.obs that correspond to categorical data. These covariates can be added in\naddition to the batch covariate and are also treated as nuisance factors\n(i.e., the model tries to minimize their effects on the latent space).\nThus, these should not be used for biologically-relevant factors that you do _not_ want to correct for.\nImportant: the order of the categorical covariates matters and should match the order of the covariates in the trained reference model.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_obs_continuous_covariate",
+          "description" : "Keys in adata.obs that correspond to continuous data. These covariates can be added in\naddition to the batch covariate and are also treated as nuisance factors\n(i.e., the model tries to minimize their effects on the latent space). Thus, these should not be\nused for biologically-relevant factors that you do _not_ want to correct for.\nImportant: the order of the continuous covariates matters and should match the order of the covariates in the trained reference model.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--sanitize_ensembl_ids",
+          "description" : "Whether to sanitize ensembl ids by removing version numbers.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Reference",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--reference",
+          "alternatives" : [
+            "-r"
+          ],
+          "description" : "Path to the directory with reference model or a web link.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_class",
+          "description" : "For legacy models; the type of model (where the type of model was not saved with it; e.g. when they were generated with scvi-tools versions < 0.15).",
+          "example" : [
+            "SCANVI"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--model_output",
+          "description" : "Output directory for model",
+          "default" : [
+            "model"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_output",
+          "description" : "In which .obsm slot to store the resulting integrated embedding.",
+          "default" : [
+            "X_integrated_scanvi"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_output_predictions",
+          "description" : "In which .obs slot to store the resulting label predictions. Only relevant if a scANVI model was provided.",
+          "default" : [
+            "scanvi_pred"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_output_probabilities",
+          "description" : "In which .obs slot to store the probabilities of the label predictions. Only relevant if a scANVI model was provided.",
+          "default" : [
+            "scanvi_proba"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Early stopping arguments",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--early_stopping",
+          "description" : "Whether to perform early stopping with respect to the validation set.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--early_stopping_monitor",
+          "description" : "Metric logged during validation set epoch.",
+          "default" : [
+            "elbo_validation"
+          ],
+          "required" : false,
+          "choices" : [
+            "elbo_validation",
+            "reconstruction_loss_validation",
+            "kl_local_validation"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--early_stopping_patience",
+          "description" : "Number of validation epochs with no improvement after which training will be stopped.",
+          "default" : [
+            45
+          ],
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--early_stopping_min_delta",
+          "description" : "Minimum change in the monitored quantity to qualify as an improvement, i.e. an absolute change of less than min_delta, will count as no improvement.",
+          "default" : [
+            0.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Learning parameters",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--max_epochs",
+          "description" : "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--reduce_lr_on_plateau",
+          "description" : "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--lr_factor",
+          "description" : "Factor to reduce learning rate.",
+          "default" : [
+            0.6
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--lr_patience",
+          "description" : "Number of epochs with no improvement after which learning rate will be reduced.",
+          "default" : [
+            30.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/set_var_index.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Performs reference mapping with scArches",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/annotation_test_data/scanvi_model"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/annotation_test_data/scanvi_covariate_model"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/annotation_test_data/scvi_model"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_mms.h5mu"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/HLCA_reference_model/HLCA_reference_model.zip"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "highmem",
+          "highcpu",
+          "highdisk",
+          "gpu"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "pytorch/pytorch:2.9.1-cuda12.8-cudnn9-runtime",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scvi-tools[cuda]~=1.4.1"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/integrate/scarches/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/integrate/scarches",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+import mudata
+from anndata import AnnData
+import scvi
+import pandas as pd
+import numpy as np
+import torch
+from tempfile import TemporaryDirectory
+from pathlib import Path
+import pickle
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'layer': $( if [ ! -z ${VIASH_PAR_LAYER+x} ]; then echo "r'${VIASH_PAR_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_obs_batch': $( if [ ! -z ${VIASH_PAR_INPUT_OBS_BATCH+x} ]; then echo "r'${VIASH_PAR_INPUT_OBS_BATCH//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_obs_label': $( if [ ! -z ${VIASH_PAR_INPUT_OBS_LABEL+x} ]; then echo "r'${VIASH_PAR_INPUT_OBS_LABEL//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_var_gene_names': $( if [ ! -z ${VIASH_PAR_INPUT_VAR_GENE_NAMES+x} ]; then echo "r'${VIASH_PAR_INPUT_VAR_GENE_NAMES//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_obs_size_factor': $( if [ ! -z ${VIASH_PAR_INPUT_OBS_SIZE_FACTOR+x} ]; then echo "r'${VIASH_PAR_INPUT_OBS_SIZE_FACTOR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_obs_categorical_covariate': $( if [ ! -z ${VIASH_PAR_INPUT_OBS_CATEGORICAL_COVARIATE+x} ]; then echo "r'${VIASH_PAR_INPUT_OBS_CATEGORICAL_COVARIATE//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'input_obs_continuous_covariate': $( if [ ! -z ${VIASH_PAR_INPUT_OBS_CONTINUOUS_COVARIATE+x} ]; then echo "r'${VIASH_PAR_INPUT_OBS_CONTINUOUS_COVARIATE//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'sanitize_ensembl_ids': $( if [ ! -z ${VIASH_PAR_SANITIZE_ENSEMBL_IDS+x} ]; then echo "r'${VIASH_PAR_SANITIZE_ENSEMBL_IDS//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'reference': $( if [ ! -z ${VIASH_PAR_REFERENCE+x} ]; then echo "r'${VIASH_PAR_REFERENCE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'reference_class': $( if [ ! -z ${VIASH_PAR_REFERENCE_CLASS+x} ]; then echo "r'${VIASH_PAR_REFERENCE_CLASS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'model_output': $( if [ ! -z ${VIASH_PAR_MODEL_OUTPUT+x} ]; then echo "r'${VIASH_PAR_MODEL_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_output': $( if [ ! -z ${VIASH_PAR_OBSM_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OBSM_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_output_predictions': $( if [ ! -z ${VIASH_PAR_OBS_OUTPUT_PREDICTIONS+x} ]; then echo "r'${VIASH_PAR_OBS_OUTPUT_PREDICTIONS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_output_probabilities': $( if [ ! -z ${VIASH_PAR_OBS_OUTPUT_PROBABILITIES+x} ]; then echo "r'${VIASH_PAR_OBS_OUTPUT_PROBABILITIES//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'early_stopping': $( if [ ! -z ${VIASH_PAR_EARLY_STOPPING+x} ]; then echo "r'${VIASH_PAR_EARLY_STOPPING//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'early_stopping_monitor': $( if [ ! -z ${VIASH_PAR_EARLY_STOPPING_MONITOR+x} ]; then echo "r'${VIASH_PAR_EARLY_STOPPING_MONITOR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'early_stopping_patience': $( if [ ! -z ${VIASH_PAR_EARLY_STOPPING_PATIENCE+x} ]; then echo "int(r'${VIASH_PAR_EARLY_STOPPING_PATIENCE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'early_stopping_min_delta': $( if [ ! -z ${VIASH_PAR_EARLY_STOPPING_MIN_DELTA+x} ]; then echo "float(r'${VIASH_PAR_EARLY_STOPPING_MIN_DELTA//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'max_epochs': $( if [ ! -z ${VIASH_PAR_MAX_EPOCHS+x} ]; then echo "int(r'${VIASH_PAR_MAX_EPOCHS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'reduce_lr_on_plateau': $( if [ ! -z ${VIASH_PAR_REDUCE_LR_ON_PLATEAU+x} ]; then echo "r'${VIASH_PAR_REDUCE_LR_ON_PLATEAU//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'lr_factor': $( if [ ! -z ${VIASH_PAR_LR_FACTOR+x} ]; then echo "float(r'${VIASH_PAR_LR_FACTOR//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'lr_patience': $( if [ ! -z ${VIASH_PAR_LR_PATIENCE+x} ]; then echo "float(r'${VIASH_PAR_LR_PATIENCE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from set_var_index import set_var_index
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+
+def _read_model_name_from_registry(model_path) -> str:
+    """Read registry with information about the model, return the model name"""
+    registry = scvi.model.base.BaseModelClass.load_registry(model_path)
+    return registry["model_name"]
+
+
+def _detect_base_model(model_path):
+    """Read from the model's file which scvi_tools model it contains"""
+    return getattr(scvi.model, _read_model_name_from_registry(model_path))
+
+
+def _validate_obs_metadata_params(model_registry, model_name):
+    """
+    Validates .obs metadata par variables that are required by scvi-tools models.
+
+    This function checks that necessary .obs metadata field names (batch, labels, size factors)
+    specified in the model registry have the required corresponding parameters provided in the input.
+
+    Parameters
+    ----------
+    model_registry : dict
+        Dictionary containing model configuration and requirements.
+    model_name : str
+        Name of the model being validated.
+    """
+
+    # Maps the keys of the model registry to their corresponding par keys containing the .obs metadata
+    registry_par_mapper = {
+        "batch_key": "input_obs_batch",
+        "labels_key": "input_obs_label",
+        "size_factor_key": "input_obs_size_factor",
+        "categorical_covariate_keys": "input_obs_categorical_covariate",
+        "continuous_covariate_keys": "input_obs_continuous_covariate",
+    }
+
+    for registry_key, key in registry_par_mapper.items():
+        model_registry_key = model_registry.get(registry_key)
+        par_key = par[key]
+
+        if model_registry_key and not par_key:
+            if key != "input_obs_label" or not model_registry.get("unlabeled_category"):
+                raise ValueError(
+                    f"The provided {model_name} model requires \\`--{key}\\` to be provided."
+                )
+
+        elif par_key and not model_registry_key:
+            logger.warning(
+                f"\\`--{key}\\` was provided but is not used in the provided {model_name} model."
+            )
+
+        elif model_registry_key and par_key:
+            if key in [
+                "input_obs_categorical_covariate",
+                "input_obs_continuous_covariate",
+            ]:
+                if len(par_key) != len(model_registry_key):
+                    raise ValueError(
+                        f"The number of provided covariates in \\`--{key}\\` ({par_key}) does not match the number of covariates used in the provided model ({model_registry_key})."
+                    )
+
+
+def _align_query_with_registry(adata_query, model_path):
+    """
+    Creates a qeury AnnData object with the expected structure and metadata fields that are aligned with the pre-trained reference model.
+
+    Parameters
+    ----------
+    adata_query : AnnData
+        The query AnnData object to be aligned with the model structure.
+    model_path : str
+        Path to the directory containing the pre-trained model.
+
+    Returns
+    -------
+    AnnData
+        A new AnnData object with structure and metadata aligned to match the
+        requirements of the pre-trained model.
+    """
+
+    model = _detect_base_model(model_path)
+    model_name = _read_model_name_from_registry(model_path)
+    model_registry = model.load_registry(model_path)["setup_args"]
+    _validate_obs_metadata_params(model_registry, model_name)
+
+    # Sanitize gene names and set as index of the AnnData object
+    # all scArches VAE models expect gene names to be in the .var index
+    adata_query = set_var_index(
+        adata_query, par["input_var_gene_names"], par["sanitize_ensembl_ids"]
+    )
+
+    # align layer
+    query_layer = (
+        adata_query.X if not par["layer"] else adata_query.layers[par["layer"]]
+    )
+    var_index = adata_query.var_names.tolist()
+    obs_index = adata_query.obs_names.tolist()
+
+    # align observations
+    query_obs = {}
+
+    ## batch_key, size_factor_key
+    simple_mappings = {
+        "batch_key": "input_obs_batch",  # relevant for AUTOZI, LinearSCVI, PEAKVI, SCANVI, SCVI, TOTALVI, MULTIVI, JaxSCVI
+        "size_factor_key": "input_obs_size_factor",  # relevant for SCANVI, SCVI, TOTALVI, MULTIVI
+    }
+
+    for registry_key, par_key in simple_mappings.items():
+        if model_registry.get(registry_key):
+            query_obs[model_registry[registry_key]] = adata_query.obs[
+                par[par_key]
+            ].tolist()
+
+    ## labels-key, relevant for AUTOZI, CondSCVI, LinearSCVI, PEAKVI, SCANVI, SCVI
+    if model_registry.get("labels_key"):
+        if par["input_obs_label"]:
+            query_obs[model_registry["labels_key"]] = adata_query.obs[
+                par["input_obs_label"]
+            ].tolist()
+        else:
+            adata_query.obs[model_registry["labels_key"]] = model_registry[
+                "unlabeled_category"
+            ]
+            query_obs[model_registry["labels_key"]] = adata_query.obs[
+                model_registry["labels_key"]
+            ].tolist()
+
+    ## covariate keys
+    covariate_mappings = {
+        "categorical_covariate_keys": "input_obs_categorical_covariate",
+        "continuous_covariate_keys": "input_obs_continuous_covariate",
+    }
+    for registry_key, par_key in covariate_mappings.items():
+        if model_registry.get(registry_key):
+            for covariate, par_covariate in zip(
+                model_registry[registry_key], par[par_key]
+            ):
+                query_obs[covariate] = adata_query.obs[par_covariate].tolist()
+
+    obs = pd.DataFrame(query_obs, index=obs_index)
+    var = pd.DataFrame(index=var_index)
+
+    aligned_query_anndata = AnnData(X=query_layer, obs=obs, var=var)
+    if model_registry["layer"]:
+        aligned_query_anndata.layers[model_registry["layer"]] = query_layer
+
+    return aligned_query_anndata
+
+
+def map_to_existing_reference(adata_query, model_path, check_val_every_n_epoch=1):
+    """
+    A function to map the query data to the reference atlas.
+
+    Input:
+        * adata_query: An AnnData object with the query
+        * model_path: The reference model directory
+
+    Output:
+        * vae_query: the trained scvi_tools model
+        * adata_query: The AnnData object with the query preprocessed for the mapping to the reference
+    """
+    model = _detect_base_model(model_path)
+
+    # Keys of the AnnData query object need to match the exact keys in the reference model registry
+
+    aligned_adata_query = _align_query_with_registry(adata_query, model_path)
+
+    try:
+        model.prepare_query_anndata(aligned_adata_query, model_path)
+    except ValueError:
+        logger.warning(
+            "ValueError thrown when preparing adata for mapping. Clearing .varm field to prevent it"
+        )
+        aligned_adata_query.varm.clear()
+        try:
+            model.prepare_query_anndata(aligned_adata_query, model_path)
+        except ValueError:
+            raise ValueError(
+                f"Could not perform model.prepare_model_anndata, likely because the model was trained with different var names then were found in the index. \\\\n\\\\nmodel var_names: {model.prepare_query_anndata(aligned_adata_query, model_path, return_reference_var_names=True).tolist()} \\\\n\\\\nquery data var_names: {aligned_adata_query.var_names.tolist()}  "
+            )
+
+    try:
+        # Load query data into the model
+        vae_query = model.load_query_data(
+            aligned_adata_query, model_path, freeze_dropout=True
+        )
+    except KeyError:
+        # Older models do not have the 'setup_method_name' key saved with the model.
+        # Assume these were generated from an anndata object.
+        model_torch = torch.load(
+            f"{model_path}/model.pt", weights_only=False, map_location="cpu"
+        )
+        model_torch["attr_dict"]["registry_"]["setup_method_name"] = "setup_anndata"
+
+        with TemporaryDirectory(dir=meta["temp_dir"]) as tempdir:
+            temp_file_name = Path(tempdir) / "model.pt"
+            torch.save(model_torch, temp_file_name)
+            del model_torch
+            vae_query = model.load_query_data(
+                aligned_adata_query, tempdir, freeze_dropout=True
+            )
+
+    # Train scArches model for query mapping
+    vae_query.train(
+        max_epochs=par["max_epochs"],
+        early_stopping=par["early_stopping"],
+        early_stopping_monitor=par["early_stopping_monitor"],
+        early_stopping_patience=par["early_stopping_patience"],
+        early_stopping_min_delta=par["early_stopping_min_delta"],
+        check_val_every_n_epoch=check_val_every_n_epoch,
+        accelerator="auto",
+    )
+
+    return vae_query
+
+
+def _convert_object_dtypes_to_strings(adata):
+    """Convert object dtypes in .var and .obs to string to prevent error when saving file"""
+
+    def convert_cols(df):
+        object_cols = df.columns[df.dtypes == "object"]
+        for col in object_cols:
+            df[col] = df[col].astype(str)
+        return df
+
+    adata.var = convert_cols(adata.var)
+    adata.obs = convert_cols(adata.obs)
+
+    return adata
+
+
+def _get_model_path(model_path: str):
+    """Obtain path to the directory with reference model. If the proposed \\`model_path\\` is a .zip archive, unzip it. If nesessary, convert model to the new format
+
+    Parameters
+    ----------
+    model_path : str
+        Path to a directory, where to search for the model or to a zip file containing the model
+
+    Returns
+    -------
+    Path to a directory with reference model in format of scvi-tools>=0.15
+    """
+    import os
+    import zipfile
+    import tempfile
+    from pathlib import Path
+
+    if os.path.isdir(model_path) and "model.pt" in os.listdir(model_path):
+        # Probably, the \\`model_path\\` already contains model in the output format of scvi-tools>=0.15
+        return model_path
+
+    # The model either has old format or is a zip file downloaded from Zenodo
+    new_directory = Path(tempfile.TemporaryDirectory().name)
+
+    if zipfile.is_zipfile(model_path):
+        with zipfile.ZipFile(model_path) as archive:
+            archive.extractall(new_directory)
+            model_dir = next(new_directory.glob("**/*.pt")).parent
+
+    else:
+        model_dir = next(Path(model_path).glob("**/*.pt")).parent
+
+    if "model_params.pt" in os.listdir(model_dir):
+        try:
+            # The current approach is to store the class with the model in a registry
+            model_class = _detect_base_model(model_dir)
+        except ValueError:
+            # Failed to load the registry, try
+            with (model_dir / "attr.pkl").open("rb") as open_file:
+                attrs = pickle.load(open_file)
+            try:
+                model_name = attrs["registry_"]["model_name"]
+            except KeyError:
+                # With older scvi-tools models, the model name was not stored with the model...
+                model_name = par["reference_class"]
+                if not model_name:
+                    raise ValueError(
+                        "Could not load reference model. This might be because it is a legacy model for which the model type was not saved with the model. Please provide it manually using the 'reference_class' argument."
+                    )
+                if not hasattr(scvi.model, model_name):
+                    raise ValueError(
+                        f"Requested to load legacy model using type {model_name}, but such a class does not exist in \\`scvi.models\\`."
+                    )
+            model_class = getattr(scvi.model, model_name)
+
+        # The model is in the \\`directory\\`, but it was generated with scvi-tools<0.15
+        # TODO: for new references (that could not be SCANVI based), we need to check the base class somehow. Reading registry does not work with models generated by scvi-tools<0.15
+        # Here I assume that the reference model is for HLCA and thus is SCANVI based
+        converted_model_path = os.path.join(model_dir, "converted")
+        model_class.convert_legacy_save(model_dir, converted_model_path)
+        return converted_model_path
+
+    elif "model.pt" in os.listdir(model_dir):
+        # Archive contained model in the new format, so just return the directory
+        return model_dir
+
+    else:
+        raise ValueError(
+            "Cannot find model in the provided reference path. Please, provide a path or a link to the directory with reference model. For HLCA use https://zenodo.org/record/6337966/files/HLCA_reference_model.zip"
+        )
+
+
+def main():
+    logger.info("Reading %s, modality %s", par["input"], par["modality"])
+    adata = mudata.read_h5ad(par["input"].strip(), mod=par["modality"])
+    adata_query = adata.copy()
+
+    model_path = _get_model_path(par["reference"])
+    vae_query = map_to_existing_reference(adata_query, model_path=model_path)
+    model_name = _read_model_name_from_registry(model_path)
+
+    # Save info about the used model
+    adata.uns["integration_method"] = model_name
+
+    logger.info("Trying to write latent representation")
+    output_key = par["obsm_output"].format(model_name=model_name)
+    adata.obsm[output_key] = vae_query.get_latent_representation()
+
+    # In addition to integrating the data, scANVI also performs label annotations
+    if model_name == "SCANVI":
+        adata.obs[par["obs_output_predictions"]] = vae_query.predict()
+        adata.obs[par["obs_output_probabilities"]] = np.max(
+            vae_query.predict(soft=True), axis=1
+        )
+
+    logger.info("Converting dtypes")
+    adata = _convert_object_dtypes_to_strings(adata)
+
+    logger.info(
+        "Saving h5mu file to %s with compression %s",
+        par["output"],
+        par["output_compression"],
+    )
+    write_h5ad_to_h5mu_with_compression(
+        par["output"], par["input"], par["modality"], adata, par["output_compression"]
+    )
+
+    logger.info("Saving model")
+    vae_query.save(par["model_output"], overwrite=True)
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/integrate/scarches",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "highmem",
+    "highcpu",
+    "highdisk",
+    "gpu"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/nextflow.config
new file mode 100644
index 0000000..69363e5
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'integrate/scarches'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Performs reference mapping with scArches'
+  author = 'Vladimir Shitov, Dorien Roosen'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/nextflow_schema.json
new file mode 100644
index 0000000..1936ec2
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/nextflow_schema.json
@@ -0,0 +1,239 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "scarches",
+  "description": "Performs reference mapping with scArches",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "Arguments related to the input (query) dataset",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file to use as a query",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`. "
+        },
+        "layer": {
+          "type": "string",
+          "description": "Layer to be used for scArches, if .X is not to be used.",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "input_obs_batch": {
+          "type": "string",
+          "description": "Name of the .obs column with batch information.",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "input_obs_label": {
+          "type": "string",
+          "description": "Name of the .obs column with celltype information.",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "input_var_gene_names": {
+          "type": "string",
+          "description": "Name of the .var column with gene names, if the var .index is not to be used.",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "input_obs_size_factor": {
+          "type": "string",
+          "description": "Key in adata.obs for size factor information",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "input_obs_categorical_covariate": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys in adata.obs that correspond to categorical data",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "input_obs_continuous_covariate": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys in adata.obs that correspond to continuous data",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "sanitize_ensembl_ids": {
+          "type": "boolean",
+          "description": "Whether to sanitize ensembl ids by removing version numbers.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output\"`, direction: `output`. ",
+          "default": "$id.$key.output"
+        },
+        "model_output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output directory for model",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"model\"`, direction: `output`. ",
+          "default": "model"
+        },
+        "obsm_output": {
+          "type": "string",
+          "description": "In which .obsm slot to store the resulting integrated embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_integrated_scanvi\"`. ",
+          "default": "X_integrated_scanvi"
+        },
+        "obs_output_predictions": {
+          "type": "string",
+          "description": "In which .obs slot to store the resulting label predictions",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scanvi_pred\"`. ",
+          "default": "scanvi_pred"
+        },
+        "obs_output_probabilities": {
+          "type": "string",
+          "description": "In which .obs slot to store the probabilities of the label predictions",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scanvi_proba\"`. ",
+          "default": "scanvi_proba"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "reference": {
+      "title": "Reference",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "reference": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the directory with reference model or a web link.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`. "
+        },
+        "reference_class": {
+          "type": "string",
+          "description": "For legacy models; the type of model (where the type of model was not saved with it; e.g",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"SCANVI\"`. "
+        }
+      }
+    },
+    "early stopping arguments": {
+      "title": "Early stopping arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "early_stopping": {
+          "type": "boolean",
+          "description": "Whether to perform early stopping with respect to the validation set.",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        },
+        "early_stopping_monitor": {
+          "type": "string",
+          "description": "Metric logged during validation set epoch.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"elbo_validation\"`, choices: ``elbo_validation`, `reconstruction_loss_validation`, `kl_local_validation``. ",
+          "enum": [
+            "elbo_validation",
+            "reconstruction_loss_validation",
+            "kl_local_validation"
+          ],
+          "default": "elbo_validation"
+        },
+        "early_stopping_patience": {
+          "type": "integer",
+          "description": "Number of validation epochs with no improvement after which training will be stopped.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `45`. ",
+          "default": 45
+        },
+        "early_stopping_min_delta": {
+          "type": "number",
+          "description": "Minimum change in the monitored quantity to qualify as an improvement, i.e",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.0`. ",
+          "default": 0.0
+        }
+      }
+    },
+    "learning parameters": {
+      "title": "Learning parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "max_epochs": {
+          "type": "integer",
+          "description": "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest.",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "reduce_lr_on_plateau": {
+          "type": "boolean",
+          "description": "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "lr_factor": {
+          "type": "number",
+          "description": "Factor to reduce learning rate.",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.6`. ",
+          "default": 0.6
+        },
+        "lr_patience": {
+          "type": "number",
+          "description": "Number of epochs with no improvement after which learning rate will be reduced.",
+          "help_text": "Type: `double`, multiple: `False`, default: `30.0`. ",
+          "default": 30.0
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/reference"
+    },
+    {
+      "$ref": "#/$defs/early stopping arguments"
+    },
+    {
+      "$ref": "#/$defs/learning parameters"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/set_var_index.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/set_var_index.py
new file mode 100644
index 0000000..1dd7e0d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/set_var_index.py
@@ -0,0 +1,68 @@
+import anndata as ad
+
+
+def strip_version_number(gene_series: list[str]) -> list[str]:
+    """Sanitize ensemble ID's by removing version numbers.
+
+    Parameters
+    ----------
+    gene_series : list[str]
+        List of ensemble ID's to sanitize.
+
+    Returns
+    -------
+    list[str]
+        List of sanitized ensemble ID's.
+    """
+
+    # Convert to string type to handle Categorical series
+    gene_series = gene_series.astype(str)
+
+    # Pattern matches Ensembl IDs: starts with ENS, followed by any characters,
+    # then an eleven digit number, optionally followed by .version_number
+    ensembl_pattern = r"^(ENS.*\d{11})(?:\.\d+)?$"
+    ensembl_mask = gene_series.str.match(ensembl_pattern)
+
+    sanitized = gene_series.where(
+        ~ensembl_mask, gene_series.str.extract(ensembl_pattern)[0]
+    )
+
+    return sanitized
+
+
+def set_var_index(
+    adata: ad.AnnData, var_name: str | None = None, sanitize_ensembl_ids: bool = True
+) -> ad.AnnData:
+    """Sanitize gene names (optional) and set the index of the .var DataFrame.
+
+    Parameters
+    ----------
+    adata : AnnData
+        Annotated data object
+    var_name : str | None
+        Name of the column in `adata.var` that contains the gene names, if None, the existing index will be sanitized but not replaced.
+    sanitize_ensembl_ids : bool
+        Whether to sanitize gene names by removing version numbers.
+
+    Returns
+    -------
+    AnnData
+        Copy of `adata` with optionally sanitized and replaced index
+    """
+    gene_names = adata.var[var_name] if var_name else adata.var.index.to_series()
+
+    if sanitize_ensembl_ids:
+        ori_gene_names = len(gene_names)
+        gene_names = strip_version_number(gene_names)
+        sanitized_gene_names = len(set(gene_names))
+
+        assert ori_gene_names == sanitized_gene_names, (
+            "Sanitizing gene names resulted in duplicated gene names.\n"
+            "Please ensure unique gene names before proceeding.\n"
+            "Please make sure --var_gene_names contains ensembl IDs (not gene symbols) "
+            "when --sanitize_ensembl_ids is set to True."
+        )
+
+    adata.var.index = gene_names.astype(str)
+
+    return adata
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scarches/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/.config.vsh.yaml
new file mode 100644
index 0000000..5824b70
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/.config.vsh.yaml
@@ -0,0 +1,648 @@
+name: "scvi"
+namespace: "integrate"
+version: "v4.0.4"
+authors:
+- name: "Malte D. Luecken"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "malte.luecken@helmholtz-muenchen.de"
+      github: "LuckyMD"
+      orcid: "0000-0001-7464-7921"
+      linkedin: "malte-l%C3%BCcken-b8b21049"
+      twitter: "MDLuecken"
+    organizations:
+    - name: "Helmholtz Munich"
+      href: "https://www.helmholtz-munich.de"
+      role: "Group Leader"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Matthias Beyens"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    links:
+      github: "MatthiasBeyens"
+      orcid: "0000-0003-3304-0706"
+      email: "matthias.beyens@gmail.com"
+      linkedin: "mbeyens"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_layer"
+    description: "Input layer to use. If None, X is used"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_batch"
+    description: "Column name discriminating between your batches."
+    info: null
+    default:
+    - "sample_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_gene_names"
+    description: ".var column containing gene names. By default, use the index."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_input"
+    description: ".var column containing highly variable genes. By default, do not\
+      \ subset genes."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_labels"
+    description: "Key in adata.obs for label information. Categories will automatically\
+      \ be \nconverted into integer categories and saved to adata.obs['_scvi_labels'].\n\
+      If None, assigns the same label to all the data.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_size_factor"
+    description: "Key in adata.obs for size factor information. Instead of using library\
+      \ size as a size factor,\nthe provided size factor column will be used as offset\
+      \ in the mean of the likelihood.\nAssumed to be on linear scale.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_categorical_covariate"
+    description: "Keys in adata.obs that correspond to categorical data. These covariates\
+      \ can be added in\naddition to the batch covariate and are also treated as nuisance\
+      \ factors\n(i.e., the model tries to minimize their effects on the latent space).\n\
+      Thus, these should not be used for biologically-relevant factors that you do\
+      \ _not_ want to correct for.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_continuous_covariate"
+    description: "Keys in adata.obs that correspond to continuous data. These covariates\
+      \ can be added in\naddition to the batch covariate and are also treated as nuisance\
+      \ factors\n(i.e., the model tries to minimize their effects on the latent space).\
+      \ Thus, these should not be\nused for biologically-relevant factors that you\
+      \ do _not_ want to correct for.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--sanitize_ensembl_ids"
+    description: "Whether to sanitize ensembl ids by removing version numbers."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_model"
+    description: "Folder where the state of the trained model will be saved to."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_output"
+    description: "In which .obsm slot to store the resulting integrated embedding."
+    info: null
+    default:
+    - "X_scvi_integrated"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "SCVI options"
+  arguments:
+  - type: "integer"
+    name: "--n_hidden_nodes"
+    description: "Number of nodes per hidden layer."
+    info: null
+    default:
+    - 128
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--n_dimensions_latent_space"
+    description: "Dimensionality of the latent space."
+    info: null
+    default:
+    - 30
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--n_hidden_layers"
+    description: "Number of hidden layers used for encoder and decoder neural-networks."
+    info: null
+    default:
+    - 2
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--dropout_rate"
+    description: "Dropout rate for the neural networks."
+    info: null
+    default:
+    - 0.1
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--dispersion"
+    description: "Set the behavior for the dispersion for negative binomial distributions:\n\
+      - gene: dispersion parameter of negative binomial is constant per gene across\
+      \ cells\n- gene-batch: dispersion can differ between different batches\n- gene-label:\
+      \ dispersion can differ between different labels\n- gene-cell:  dispersion can\
+      \ differ for every gene in every cell\n"
+    info: null
+    default:
+    - "gene"
+    required: false
+    choices:
+    - "gene"
+    - "gene-batch"
+    - "gene-label"
+    - "gene-cell"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--gene_likelihood"
+    description: "Model used to generate the expression data from a count-based likelihood\
+      \ distribution.\n- nb: Negative binomial distribution\n- zinb: Zero-inflated\
+      \ negative binomial distribution\n- poisson: Poisson distribution\n"
+    info: null
+    default:
+    - "nb"
+    required: false
+    choices:
+    - "nb"
+    - "zinb"
+    - "poisson"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Variational auto-encoder model options"
+  arguments:
+  - type: "string"
+    name: "--use_layer_normalization"
+    description: "Neural networks for which to enable layer normalization. \n"
+    info: null
+    default:
+    - "both"
+    required: false
+    choices:
+    - "encoder"
+    - "decoder"
+    - "none"
+    - "both"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--use_batch_normalization"
+    description: "Neural networks for which to enable batch normalization. \n"
+    info: null
+    default:
+    - "none"
+    required: false
+    choices:
+    - "encoder"
+    - "decoder"
+    - "none"
+    - "both"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_false"
+    name: "--encode_covariates"
+    description: "Whether to concatenate covariates to expression in encoder"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--deeply_inject_covariates"
+    description: "Whether to concatenate covariates into output of hidden layers in\
+      \ encoder/decoder. \nThis option only applies when n_layers > 1. The covariates\
+      \ are concatenated to\nthe input of subsequent hidden layers.\n"
+    info: null
+    direction: "input"
+  - type: "boolean_true"
+    name: "--use_observed_lib_size"
+    description: "Use observed library size for RNA as scaling factor in mean of conditional\
+      \ distribution.\n"
+    info: null
+    direction: "input"
+- name: "Early stopping arguments"
+  arguments:
+  - type: "boolean"
+    name: "--early_stopping"
+    description: "Whether to perform early stopping with respect to the validation\
+      \ set."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--early_stopping_monitor"
+    description: "Metric logged during validation set epoch."
+    info: null
+    default:
+    - "elbo_validation"
+    required: false
+    choices:
+    - "elbo_validation"
+    - "reconstruction_loss_validation"
+    - "kl_local_validation"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--early_stopping_patience"
+    description: "Number of validation epochs with no improvement after which training\
+      \ will be stopped."
+    info: null
+    default:
+    - 45
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--early_stopping_min_delta"
+    description: "Minimum change in the monitored quantity to qualify as an improvement,\
+      \ i.e. an absolute change of less than min_delta, will count as no improvement."
+    info: null
+    default:
+    - 0.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Learning parameters"
+  arguments:
+  - type: "integer"
+    name: "--max_epochs"
+    description: "Number of passes through the dataset, defaults to (20000 / number\
+      \ of cells) * 400 or 400; whichever is smallest."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--reduce_lr_on_plateau"
+    description: "Whether to monitor validation loss and reduce learning rate when\
+      \ validation set `lr_scheduler_metric` plateaus."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--lr_factor"
+    description: "Factor to reduce learning rate."
+    info: null
+    default:
+    - 0.6
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--lr_patience"
+    description: "Number of epochs with no improvement after which learning rate will\
+      \ be reduced."
+    info: null
+    default:
+    - 30.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Data validition"
+  arguments:
+  - type: "integer"
+    name: "--n_obs_min_count"
+    description: "Minimum number of cells threshold ensuring that every obs_batch\
+      \ category has sufficient observations (cells) for model training."
+    info: null
+    default:
+    - 0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--n_var_min_count"
+    description: "Minimum number of genes threshold ensuring that every var_input\
+      \ filter has sufficient observations (genes) for model training."
+    info: null
+    default:
+    - 0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "subset_vars.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "set_var_index.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Performs scvi integration as done in the human lung cell atlas https://github.com/LungCellAtlas/HLCA"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3_mms.h5mu"
+- type: "file"
+  path: "TS_Blood_filtered.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "midcpu"
+    - "midmem"
+    - "gpu"
+    - "highdisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "pytorch/pytorch:2.9.1-cuda12.8-cudnn9-runtime"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  - type: "python"
+    user: false
+    packages:
+    - "scvi-tools[cuda]~=1.4.1"
+    upgrade: true
+  test_setup:
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/integrate/scvi/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/integrate/scvi"
+  executable: "target/nextflow/integrate/scvi/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/main.nf
new file mode 100644
index 0000000..6597b57
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/main.nf
@@ -0,0 +1,4516 @@
+// scvi v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Malte D. Luecken (author)
+//  * Dries Schaumont (maintainer)
+//  * Matthias Beyens (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "scvi",
+  "namespace" : "integrate",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Malte D. Luecken",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "malte.luecken@helmholtz-muenchen.de",
+          "github" : "LuckyMD",
+          "orcid" : "0000-0001-7464-7921",
+          "linkedin" : "malte-l%C3%BCcken-b8b21049",
+          "twitter" : "MDLuecken"
+        },
+        "organizations" : [
+          {
+            "name" : "Helmholtz Munich",
+            "href" : "https://www.helmholtz-munich.de",
+            "role" : "Group Leader"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Matthias Beyens",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "links" : {
+          "github" : "MatthiasBeyens",
+          "orcid" : "0000-0003-3304-0706",
+          "email" : "matthias.beyens@gmail.com",
+          "linkedin" : "mbeyens"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_layer",
+          "description" : "Input layer to use. If None, X is used",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_batch",
+          "description" : "Column name discriminating between your batches.",
+          "default" : [
+            "sample_id"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_gene_names",
+          "description" : ".var column containing gene names. By default, use the index.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_input",
+          "description" : ".var column containing highly variable genes. By default, do not subset genes.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_labels",
+          "description" : "Key in adata.obs for label information. Categories will automatically be \nconverted into integer categories and saved to adata.obs['_scvi_labels'].\nIf None, assigns the same label to all the data.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_size_factor",
+          "description" : "Key in adata.obs for size factor information. Instead of using library size as a size factor,\nthe provided size factor column will be used as offset in the mean of the likelihood.\nAssumed to be on linear scale.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_categorical_covariate",
+          "description" : "Keys in adata.obs that correspond to categorical data. These covariates can be added in\naddition to the batch covariate and are also treated as nuisance factors\n(i.e., the model tries to minimize their effects on the latent space).\nThus, these should not be used for biologically-relevant factors that you do _not_ want to correct for.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_continuous_covariate",
+          "description" : "Keys in adata.obs that correspond to continuous data. These covariates can be added in\naddition to the batch covariate and are also treated as nuisance factors\n(i.e., the model tries to minimize their effects on the latent space). Thus, these should not be\nused for biologically-relevant factors that you do _not_ want to correct for.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--sanitize_ensembl_ids",
+          "description" : "Whether to sanitize ensembl ids by removing version numbers.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_model",
+          "description" : "Folder where the state of the trained model will be saved to.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_output",
+          "description" : "In which .obsm slot to store the resulting integrated embedding.",
+          "default" : [
+            "X_scvi_integrated"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "SCVI options",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--n_hidden_nodes",
+          "description" : "Number of nodes per hidden layer.",
+          "default" : [
+            128
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--n_dimensions_latent_space",
+          "description" : "Dimensionality of the latent space.",
+          "default" : [
+            30
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--n_hidden_layers",
+          "description" : "Number of hidden layers used for encoder and decoder neural-networks.",
+          "default" : [
+            2
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--dropout_rate",
+          "description" : "Dropout rate for the neural networks.",
+          "default" : [
+            0.1
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--dispersion",
+          "description" : "Set the behavior for the dispersion for negative binomial distributions:\n- gene: dispersion parameter of negative binomial is constant per gene across cells\n- gene-batch: dispersion can differ between different batches\n- gene-label: dispersion can differ between different labels\n- gene-cell:  dispersion can differ for every gene in every cell\n",
+          "default" : [
+            "gene"
+          ],
+          "required" : false,
+          "choices" : [
+            "gene",
+            "gene-batch",
+            "gene-label",
+            "gene-cell"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--gene_likelihood",
+          "description" : "Model used to generate the expression data from a count-based likelihood distribution.\n- nb: Negative binomial distribution\n- zinb: Zero-inflated negative binomial distribution\n- poisson: Poisson distribution\n",
+          "default" : [
+            "nb"
+          ],
+          "required" : false,
+          "choices" : [
+            "nb",
+            "zinb",
+            "poisson"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Variational auto-encoder model options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--use_layer_normalization",
+          "description" : "Neural networks for which to enable layer normalization. \n",
+          "default" : [
+            "both"
+          ],
+          "required" : false,
+          "choices" : [
+            "encoder",
+            "decoder",
+            "none",
+            "both"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--use_batch_normalization",
+          "description" : "Neural networks for which to enable batch normalization. \n",
+          "default" : [
+            "none"
+          ],
+          "required" : false,
+          "choices" : [
+            "encoder",
+            "decoder",
+            "none",
+            "both"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_false",
+          "name" : "--encode_covariates",
+          "description" : "Whether to concatenate covariates to expression in encoder",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--deeply_inject_covariates",
+          "description" : "Whether to concatenate covariates into output of hidden layers in encoder/decoder. \nThis option only applies when n_layers > 1. The covariates are concatenated to\nthe input of subsequent hidden layers.\n",
+          "direction" : "input"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--use_observed_lib_size",
+          "description" : "Use observed library size for RNA as scaling factor in mean of conditional distribution.\n",
+          "direction" : "input"
+        }
+      ]
+    },
+    {
+      "name" : "Early stopping arguments",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--early_stopping",
+          "description" : "Whether to perform early stopping with respect to the validation set.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--early_stopping_monitor",
+          "description" : "Metric logged during validation set epoch.",
+          "default" : [
+            "elbo_validation"
+          ],
+          "required" : false,
+          "choices" : [
+            "elbo_validation",
+            "reconstruction_loss_validation",
+            "kl_local_validation"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--early_stopping_patience",
+          "description" : "Number of validation epochs with no improvement after which training will be stopped.",
+          "default" : [
+            45
+          ],
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--early_stopping_min_delta",
+          "description" : "Minimum change in the monitored quantity to qualify as an improvement, i.e. an absolute change of less than min_delta, will count as no improvement.",
+          "default" : [
+            0.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Learning parameters",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--max_epochs",
+          "description" : "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--reduce_lr_on_plateau",
+          "description" : "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--lr_factor",
+          "description" : "Factor to reduce learning rate.",
+          "default" : [
+            0.6
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--lr_patience",
+          "description" : "Number of epochs with no improvement after which learning rate will be reduced.",
+          "default" : [
+            30.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Data validition",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--n_obs_min_count",
+          "description" : "Minimum number of cells threshold ensuring that every obs_batch category has sufficient observations (cells) for model training.",
+          "default" : [
+            0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--n_var_min_count",
+          "description" : "Minimum number of genes threshold ensuring that every var_input filter has sufficient observations (genes) for model training.",
+          "default" : [
+            0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/subset_vars.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/set_var_index.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Performs scvi integration as done in the human lung cell atlas https://github.com/LungCellAtlas/HLCA",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_mms.h5mu"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/annotation_test_data/TS_Blood_filtered.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "midcpu",
+          "midmem",
+          "gpu",
+          "highdisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "pytorch/pytorch:2.9.1-cuda12.8-cudnn9-runtime",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "scvi-tools[cuda]~=1.4.1"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/integrate/scvi/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/integrate/scvi",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+from scanpy._utils import check_nonnegative_integers
+import mudata
+import scvi
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_layer': $( if [ ! -z ${VIASH_PAR_INPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_INPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_batch': $( if [ ! -z ${VIASH_PAR_OBS_BATCH+x} ]; then echo "r'${VIASH_PAR_OBS_BATCH//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_gene_names': $( if [ ! -z ${VIASH_PAR_VAR_GENE_NAMES+x} ]; then echo "r'${VIASH_PAR_VAR_GENE_NAMES//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_input': $( if [ ! -z ${VIASH_PAR_VAR_INPUT+x} ]; then echo "r'${VIASH_PAR_VAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_labels': $( if [ ! -z ${VIASH_PAR_OBS_LABELS+x} ]; then echo "r'${VIASH_PAR_OBS_LABELS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_size_factor': $( if [ ! -z ${VIASH_PAR_OBS_SIZE_FACTOR+x} ]; then echo "r'${VIASH_PAR_OBS_SIZE_FACTOR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_categorical_covariate': $( if [ ! -z ${VIASH_PAR_OBS_CATEGORICAL_COVARIATE+x} ]; then echo "r'${VIASH_PAR_OBS_CATEGORICAL_COVARIATE//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'obs_continuous_covariate': $( if [ ! -z ${VIASH_PAR_OBS_CONTINUOUS_COVARIATE+x} ]; then echo "r'${VIASH_PAR_OBS_CONTINUOUS_COVARIATE//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'sanitize_ensembl_ids': $( if [ ! -z ${VIASH_PAR_SANITIZE_ENSEMBL_IDS+x} ]; then echo "r'${VIASH_PAR_SANITIZE_ENSEMBL_IDS//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_model': $( if [ ! -z ${VIASH_PAR_OUTPUT_MODEL+x} ]; then echo "r'${VIASH_PAR_OUTPUT_MODEL//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_output': $( if [ ! -z ${VIASH_PAR_OBSM_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OBSM_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'n_hidden_nodes': $( if [ ! -z ${VIASH_PAR_N_HIDDEN_NODES+x} ]; then echo "int(r'${VIASH_PAR_N_HIDDEN_NODES//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'n_dimensions_latent_space': $( if [ ! -z ${VIASH_PAR_N_DIMENSIONS_LATENT_SPACE+x} ]; then echo "int(r'${VIASH_PAR_N_DIMENSIONS_LATENT_SPACE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'n_hidden_layers': $( if [ ! -z ${VIASH_PAR_N_HIDDEN_LAYERS+x} ]; then echo "int(r'${VIASH_PAR_N_HIDDEN_LAYERS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'dropout_rate': $( if [ ! -z ${VIASH_PAR_DROPOUT_RATE+x} ]; then echo "float(r'${VIASH_PAR_DROPOUT_RATE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'dispersion': $( if [ ! -z ${VIASH_PAR_DISPERSION+x} ]; then echo "r'${VIASH_PAR_DISPERSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'gene_likelihood': $( if [ ! -z ${VIASH_PAR_GENE_LIKELIHOOD+x} ]; then echo "r'${VIASH_PAR_GENE_LIKELIHOOD//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'use_layer_normalization': $( if [ ! -z ${VIASH_PAR_USE_LAYER_NORMALIZATION+x} ]; then echo "r'${VIASH_PAR_USE_LAYER_NORMALIZATION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'use_batch_normalization': $( if [ ! -z ${VIASH_PAR_USE_BATCH_NORMALIZATION+x} ]; then echo "r'${VIASH_PAR_USE_BATCH_NORMALIZATION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'encode_covariates': $( if [ ! -z ${VIASH_PAR_ENCODE_COVARIATES+x} ]; then echo "r'${VIASH_PAR_ENCODE_COVARIATES//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'deeply_inject_covariates': $( if [ ! -z ${VIASH_PAR_DEEPLY_INJECT_COVARIATES+x} ]; then echo "r'${VIASH_PAR_DEEPLY_INJECT_COVARIATES//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'use_observed_lib_size': $( if [ ! -z ${VIASH_PAR_USE_OBSERVED_LIB_SIZE+x} ]; then echo "r'${VIASH_PAR_USE_OBSERVED_LIB_SIZE//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'early_stopping': $( if [ ! -z ${VIASH_PAR_EARLY_STOPPING+x} ]; then echo "r'${VIASH_PAR_EARLY_STOPPING//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'early_stopping_monitor': $( if [ ! -z ${VIASH_PAR_EARLY_STOPPING_MONITOR+x} ]; then echo "r'${VIASH_PAR_EARLY_STOPPING_MONITOR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'early_stopping_patience': $( if [ ! -z ${VIASH_PAR_EARLY_STOPPING_PATIENCE+x} ]; then echo "int(r'${VIASH_PAR_EARLY_STOPPING_PATIENCE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'early_stopping_min_delta': $( if [ ! -z ${VIASH_PAR_EARLY_STOPPING_MIN_DELTA+x} ]; then echo "float(r'${VIASH_PAR_EARLY_STOPPING_MIN_DELTA//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'max_epochs': $( if [ ! -z ${VIASH_PAR_MAX_EPOCHS+x} ]; then echo "int(r'${VIASH_PAR_MAX_EPOCHS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'reduce_lr_on_plateau': $( if [ ! -z ${VIASH_PAR_REDUCE_LR_ON_PLATEAU+x} ]; then echo "r'${VIASH_PAR_REDUCE_LR_ON_PLATEAU//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'lr_factor': $( if [ ! -z ${VIASH_PAR_LR_FACTOR+x} ]; then echo "float(r'${VIASH_PAR_LR_FACTOR//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'lr_patience': $( if [ ! -z ${VIASH_PAR_LR_PATIENCE+x} ]; then echo "float(r'${VIASH_PAR_LR_PATIENCE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'n_obs_min_count': $( if [ ! -z ${VIASH_PAR_N_OBS_MIN_COUNT+x} ]; then echo "int(r'${VIASH_PAR_N_OBS_MIN_COUNT//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'n_var_min_count': $( if [ ! -z ${VIASH_PAR_N_VAR_MIN_COUNT+x} ]; then echo "int(r'${VIASH_PAR_N_VAR_MIN_COUNT//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+import sys
+
+sys.path.append(meta["resources_dir"])
+
+from subset_vars import subset_vars
+from set_var_index import set_var_index
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+
+# TODO: optionally, move to qa
+# https://github.com/openpipelines-bio/openpipeline/issues/435
+def check_validity_anndata(adata, layer, obs_batch, n_obs_min_count, n_var_min_count):
+    assert check_nonnegative_integers(adata.layers[layer] if layer else adata.X), (
+        "Make sure input adata contains raw_counts"
+    )
+
+    assert len(set(adata.var_names)) == len(adata.var_names), (
+        "Dataset contains multiple genes with same gene name."
+    )
+
+    # Ensure every obs_batch category has sufficient observations
+    assert min(adata.obs[[obs_batch]].value_counts()) > n_obs_min_count, (
+        f"Anndata has fewer than {n_obs_min_count} cells."
+    )
+
+    assert adata.n_vars > n_var_min_count, (
+        f"Anndata has fewer than {n_var_min_count} genes."
+    )
+
+
+def main():
+    adata = mudata.read_h5ad(par["input"].strip(), mod=par["modality"])
+
+    if par["var_input"]:
+        # Subset to HVG
+        adata_subset = subset_vars(adata, subset_col=par["var_input"]).copy()
+    else:
+        adata_subset = adata.copy()
+
+    # Sanitize gene names and set as index of the AnnData object
+    adata_subset = set_var_index(
+        adata_subset, par["var_gene_names"], par["sanitize_ensembl_ids"]
+    )
+
+    check_validity_anndata(
+        adata_subset,
+        par["input_layer"],
+        par["obs_batch"],
+        par["n_obs_min_count"],
+        par["n_var_min_count"],
+    )
+
+    # Set up the data
+    scvi.model.SCVI.setup_anndata(
+        adata_subset,
+        batch_key=par["obs_batch"],
+        layer=par["input_layer"],
+        labels_key=par["obs_labels"],
+        size_factor_key=par["obs_size_factor"],
+        categorical_covariate_keys=par["obs_categorical_covariate"],
+        continuous_covariate_keys=par["obs_continuous_covariate"],
+    )
+
+    # Set up the model
+    vae_uns = scvi.model.SCVI(
+        adata_subset,
+        n_hidden=par["n_hidden_nodes"],
+        n_latent=par["n_dimensions_latent_space"],
+        n_layers=par["n_hidden_layers"],
+        dropout_rate=par["dropout_rate"],
+        dispersion=par["dispersion"],
+        gene_likelihood=par["gene_likelihood"],
+        use_layer_norm=par["use_layer_normalization"],
+        use_batch_norm=par["use_batch_normalization"],
+        encode_covariates=par[
+            "encode_covariates"
+        ],  # Default (True) is for better scArches performance -> maybe don't use this always?
+        deeply_inject_covariates=par[
+            "deeply_inject_covariates"
+        ],  # Default (False) for better scArches performance -> maybe don't use this always?
+        use_observed_lib_size=par[
+            "use_observed_lib_size"
+        ],  # When size_factors are not passed
+    )
+
+    plan_kwargs = {
+        "reduce_lr_on_plateau": par["reduce_lr_on_plateau"],
+        "lr_patience": par["lr_patience"],
+        "lr_factor": par["lr_factor"],
+    }
+
+    # Train the model
+    vae_uns.train(
+        max_epochs=par["max_epochs"],
+        early_stopping=par["early_stopping"],
+        early_stopping_monitor=par["early_stopping_monitor"],
+        early_stopping_patience=par["early_stopping_patience"],
+        early_stopping_min_delta=par["early_stopping_min_delta"],
+        plan_kwargs=plan_kwargs,
+        check_val_every_n_epoch=1,
+        accelerator="auto",
+    )
+    # Note: train_size=1.0 should give better results, but then can't do early_stopping on validation set
+
+    # Get the latent output
+    adata.obsm[par["obsm_output"]] = vae_uns.get_latent_representation()
+
+    write_h5ad_to_h5mu_with_compression(
+        par["output"], par["input"], par["modality"], adata, par["output_compression"]
+    )
+    if par["output_model"]:
+        vae_uns.save(par["output_model"], overwrite=True)
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/integrate/scvi",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "midcpu",
+    "midmem",
+    "gpu",
+    "highdisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/nextflow.config
new file mode 100644
index 0000000..c199698
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'integrate/scvi'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Performs scvi integration as done in the human lung cell atlas https://github.com/LungCellAtlas/HLCA'
+  author = 'Malte D. Luecken, Dries Schaumont, Matthias Beyens'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/nextflow_schema.json
new file mode 100644
index 0000000..732d768
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/nextflow_schema.json
@@ -0,0 +1,342 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "scvi",
+  "description": "Performs scvi integration as done in the human lung cell atlas https://github.com/LungCellAtlas/HLCA",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "input_layer": {
+          "type": "string",
+          "description": "Input layer to use",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_batch": {
+          "type": "string",
+          "description": "Column name discriminating between your batches.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"sample_id\"`. ",
+          "default": "sample_id"
+        },
+        "var_gene_names": {
+          "type": "string",
+          "description": ".var column containing gene names",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "var_input": {
+          "type": "string",
+          "description": ".var column containing highly variable genes",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_labels": {
+          "type": "string",
+          "description": "Key in adata.obs for label information",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_size_factor": {
+          "type": "string",
+          "description": "Key in adata.obs for size factor information",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_categorical_covariate": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys in adata.obs that correspond to categorical data",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "obs_continuous_covariate": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys in adata.obs that correspond to continuous data",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "sanitize_ensembl_ids": {
+          "type": "boolean",
+          "description": "Whether to sanitize ensembl ids by removing version numbers.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output\"`, direction: `output`. ",
+          "default": "$id.$key.output"
+        },
+        "output_model": {
+          "type": "string",
+          "format": "path",
+          "description": "Folder where the state of the trained model will be saved to.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output_model\"`, direction: `output`. ",
+          "default": "$id.$key.output_model"
+        },
+        "obsm_output": {
+          "type": "string",
+          "description": "In which .obsm slot to store the resulting integrated embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_scvi_integrated\"`. ",
+          "default": "X_scvi_integrated"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "scvi options": {
+      "title": "SCVI options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "n_hidden_nodes": {
+          "type": "integer",
+          "description": "Number of nodes per hidden layer.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `128`. ",
+          "default": 128
+        },
+        "n_dimensions_latent_space": {
+          "type": "integer",
+          "description": "Dimensionality of the latent space.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `30`. ",
+          "default": 30
+        },
+        "n_hidden_layers": {
+          "type": "integer",
+          "description": "Number of hidden layers used for encoder and decoder neural-networks.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `2`. ",
+          "default": 2
+        },
+        "dropout_rate": {
+          "type": "number",
+          "description": "Dropout rate for the neural networks.",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.1`. ",
+          "default": 0.1
+        },
+        "dispersion": {
+          "type": "string",
+          "description": "Set the behavior for the dispersion for negative binomial distributions:\n- gene: dispersion parameter of negative binomial is constant per gene across cells\n- gene-batch: dispersion can differ between different batches\n- gene-label: dispersion can differ between different labels\n- gene-cell:  dispersion can differ for every gene in every cell\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"gene\"`, choices: ``gene`, `gene-batch`, `gene-label`, `gene-cell``. ",
+          "enum": [
+            "gene",
+            "gene-batch",
+            "gene-label",
+            "gene-cell"
+          ],
+          "default": "gene"
+        },
+        "gene_likelihood": {
+          "type": "string",
+          "description": "Model used to generate the expression data from a count-based likelihood distribution.\n- nb: Negative binomial distribution\n- zinb: Zero-inflated negative binomial distribution\n- poisson: Poisson distribution\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"nb\"`, choices: ``nb`, `zinb`, `poisson``. ",
+          "enum": [
+            "nb",
+            "zinb",
+            "poisson"
+          ],
+          "default": "nb"
+        }
+      }
+    },
+    "variational auto-encoder model options": {
+      "title": "Variational auto-encoder model options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "use_layer_normalization": {
+          "type": "string",
+          "description": "Neural networks for which to enable layer normalization",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"both\"`, choices: ``encoder`, `decoder`, `none`, `both``. ",
+          "enum": [
+            "encoder",
+            "decoder",
+            "none",
+            "both"
+          ],
+          "default": "both"
+        },
+        "use_batch_normalization": {
+          "type": "string",
+          "description": "Neural networks for which to enable batch normalization",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"none\"`, choices: ``encoder`, `decoder`, `none`, `both``. ",
+          "enum": [
+            "encoder",
+            "decoder",
+            "none",
+            "both"
+          ],
+          "default": "none"
+        },
+        "encode_covariates": {
+          "type": "boolean",
+          "description": "Whether to concatenate covariates to expression in encoder",
+          "help_text": "Type: `boolean_false`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "deeply_inject_covariates": {
+          "type": "boolean",
+          "description": "Whether to concatenate covariates into output of hidden layers in encoder/decoder",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "use_observed_lib_size": {
+          "type": "boolean",
+          "description": "Use observed library size for RNA as scaling factor in mean of conditional distribution.\n",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "early stopping arguments": {
+      "title": "Early stopping arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "early_stopping": {
+          "type": "boolean",
+          "description": "Whether to perform early stopping with respect to the validation set.",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        },
+        "early_stopping_monitor": {
+          "type": "string",
+          "description": "Metric logged during validation set epoch.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"elbo_validation\"`, choices: ``elbo_validation`, `reconstruction_loss_validation`, `kl_local_validation``. ",
+          "enum": [
+            "elbo_validation",
+            "reconstruction_loss_validation",
+            "kl_local_validation"
+          ],
+          "default": "elbo_validation"
+        },
+        "early_stopping_patience": {
+          "type": "integer",
+          "description": "Number of validation epochs with no improvement after which training will be stopped.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `45`. ",
+          "default": 45
+        },
+        "early_stopping_min_delta": {
+          "type": "number",
+          "description": "Minimum change in the monitored quantity to qualify as an improvement, i.e",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.0`. ",
+          "default": 0.0
+        }
+      }
+    },
+    "learning parameters": {
+      "title": "Learning parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "max_epochs": {
+          "type": "integer",
+          "description": "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest.",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "reduce_lr_on_plateau": {
+          "type": "boolean",
+          "description": "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "lr_factor": {
+          "type": "number",
+          "description": "Factor to reduce learning rate.",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.6`. ",
+          "default": 0.6
+        },
+        "lr_patience": {
+          "type": "number",
+          "description": "Number of epochs with no improvement after which learning rate will be reduced.",
+          "help_text": "Type: `double`, multiple: `False`, default: `30.0`. ",
+          "default": 30.0
+        }
+      }
+    },
+    "data validition": {
+      "title": "Data validition",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "n_obs_min_count": {
+          "type": "integer",
+          "description": "Minimum number of cells threshold ensuring that every obs_batch category has sufficient observations (cells) for model training.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `0`. ",
+          "default": 0
+        },
+        "n_var_min_count": {
+          "type": "integer",
+          "description": "Minimum number of genes threshold ensuring that every var_input filter has sufficient observations (genes) for model training.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `0`. ",
+          "default": 0
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/scvi options"
+    },
+    {
+      "$ref": "#/$defs/variational auto-encoder model options"
+    },
+    {
+      "$ref": "#/$defs/early stopping arguments"
+    },
+    {
+      "$ref": "#/$defs/learning parameters"
+    },
+    {
+      "$ref": "#/$defs/data validition"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/set_var_index.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/set_var_index.py
new file mode 100644
index 0000000..1dd7e0d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/set_var_index.py
@@ -0,0 +1,68 @@
+import anndata as ad
+
+
+def strip_version_number(gene_series: list[str]) -> list[str]:
+    """Sanitize ensemble ID's by removing version numbers.
+
+    Parameters
+    ----------
+    gene_series : list[str]
+        List of ensemble ID's to sanitize.
+
+    Returns
+    -------
+    list[str]
+        List of sanitized ensemble ID's.
+    """
+
+    # Convert to string type to handle Categorical series
+    gene_series = gene_series.astype(str)
+
+    # Pattern matches Ensembl IDs: starts with ENS, followed by any characters,
+    # then an eleven digit number, optionally followed by .version_number
+    ensembl_pattern = r"^(ENS.*\d{11})(?:\.\d+)?$"
+    ensembl_mask = gene_series.str.match(ensembl_pattern)
+
+    sanitized = gene_series.where(
+        ~ensembl_mask, gene_series.str.extract(ensembl_pattern)[0]
+    )
+
+    return sanitized
+
+
+def set_var_index(
+    adata: ad.AnnData, var_name: str | None = None, sanitize_ensembl_ids: bool = True
+) -> ad.AnnData:
+    """Sanitize gene names (optional) and set the index of the .var DataFrame.
+
+    Parameters
+    ----------
+    adata : AnnData
+        Annotated data object
+    var_name : str | None
+        Name of the column in `adata.var` that contains the gene names, if None, the existing index will be sanitized but not replaced.
+    sanitize_ensembl_ids : bool
+        Whether to sanitize gene names by removing version numbers.
+
+    Returns
+    -------
+    AnnData
+        Copy of `adata` with optionally sanitized and replaced index
+    """
+    gene_names = adata.var[var_name] if var_name else adata.var.index.to_series()
+
+    if sanitize_ensembl_ids:
+        ori_gene_names = len(gene_names)
+        gene_names = strip_version_number(gene_names)
+        sanitized_gene_names = len(set(gene_names))
+
+        assert ori_gene_names == sanitized_gene_names, (
+            "Sanitizing gene names resulted in duplicated gene names.\n"
+            "Please ensure unique gene names before proceeding.\n"
+            "Please make sure --var_gene_names contains ensembl IDs (not gene symbols) "
+            "when --sanitize_ensembl_ids is set to True."
+        )
+
+    adata.var.index = gene_names.astype(str)
+
+    return adata
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/subset_vars.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/subset_vars.py
new file mode 100644
index 0000000..6003513
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/integrate/scvi/subset_vars.py
@@ -0,0 +1,31 @@
+def subset_vars(adata, subset_col):
+    """Subset AnnData object on highly variable genes
+
+    Parameters
+    ----------
+    adata : AnnData
+        Annotated data object
+    subset_col : str
+        Name of the boolean column in `adata.var` that contains the information if features should be used or not
+
+    Returns
+    -------
+    AnnData
+        Copy of `adata` with subsetted features
+    """
+    if subset_col not in adata.var.columns:
+        raise ValueError(
+            f"Requested to use .var column '{subset_col}' as a selection of genes, but the column is not available."
+        )
+
+    if adata.var[subset_col].dtype == "boolean":
+        assert adata.var[subset_col].isna().sum() == 0, (
+            f"The .var column `{subset_col}` contains NaN values. Can not subset data."
+        )
+        adata.var[subset_col] = adata.var[subset_col].astype("bool")
+
+    assert adata.var[subset_col].dtype == "bool", (
+        f"Expected dtype of .var column '{subset_col}' to be `bool`, but found {adata.var[subset_col].dtype}. Can not subset data."
+    )
+
+    return adata[:, adata.var[subset_col]].copy()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/.config.vsh.yaml
new file mode 100644
index 0000000..968db4d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/.config.vsh.yaml
@@ -0,0 +1,1039 @@
+name: "cellranger_multi"
+namespace: "mapping"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Angela Oliveira Pisco"
+  roles:
+  - "author"
+  info:
+    role: "Contributor"
+    links:
+      github: "aopisco"
+      orcid: "0000-0003-0142-2355"
+      linkedin: "aopisco"
+    organizations:
+    - name: "Insitro"
+      href: "https://insitro.com"
+      role: "Director of Computational Biology"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Dries De Maeyer"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Input files"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "The FASTQ files to be analyzed. FASTQ files should conform to the\
+      \ naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane\
+      \ Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Feature type-specific input files"
+  description: "Helper functionality to allow feature type-specific input files, without\
+    \ the need to specify\nlibrary_type or library_id. The library_id will be inferred\
+    \ from the input paths.\n"
+  arguments:
+  - type: "file"
+    name: "--gex_input"
+    description: "The FASTQ files to be analyzed for Gene Expression. FASTQ files\
+      \ should conform to the\nnaming conventions of bcl2fastq and mkfastq:\n`[Sample\
+      \ Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--abc_input"
+    description: "The FASTQ files to be analyzed for Antibody Capture. FASTQ files\
+      \ should conform to \nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample\
+      \ Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--cgc_input"
+    description: "The FASTQ files to be analyzed for CRISPR Guide Capture. FASTQ files\
+      \ should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample\
+      \ Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--mux_input"
+    description: "The FASTQ files to be analyzed for Multiplexing Capture. FASTQ files\
+      \ should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample\
+      \ Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--vdj_input"
+    description: "The FASTQ files to be analyzed for VDJ. FASTQ files should conform\
+      \ to the\nnaming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample\
+      \ Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--vdj_t_input"
+    description: "The FASTQ files to be analyzed for VDJ-T. FASTQ files should conform\
+      \ to the naming\nconventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample\
+      \ Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--vdj_t_gd_input"
+    description: "The FASTQ files to be analyzed for VDJ-T-GD. FASTQ files should\
+      \ conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample\
+      \ Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--vdj_b_input"
+    description: "The FASTQ files to be analyzed for VDJ-B. FASTQ files should conform\
+      \ to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample\
+      \ Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--agc_input"
+    description: "The FASTQ files to be analyzed for Antigen Capture. FASTQ files\
+      \ should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample\
+      \ Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Library arguments"
+  arguments:
+  - type: "string"
+    name: "--library_id"
+    description: "The Illumina sample name to analyze. This must exactly match the\
+      \ 'Sample Name'part\nof the FASTQ files specified in the `--input` argument.\n"
+    info: null
+    example:
+    - "mysample1"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--library_type"
+    description: "The underlying feature type of the library.\n"
+    info: null
+    example:
+    - "Gene Expression"
+    required: false
+    choices:
+    - "Gene Expression"
+    - "VDJ"
+    - "VDJ-T"
+    - "VDJ-B"
+    - "VDJ-T-GD"
+    - "Antibody Capture"
+    - "CRISPR Guide Capture"
+    - "Multiplexing Capture"
+    - "Antigen Capture"
+    - "Custom"
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--library_subsample"
+    description: "The rate at which reads from the provided FASTQ files are sampled.\n\
+      Must be strictly greater than 0 and less than or equal to 1.\n"
+    info: null
+    example:
+    - "0.5"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--library_lanes"
+    description: "Lanes associated with this sample. Defaults to using all lanes."
+    info: null
+    example:
+    - "1-4"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--library_chemistry"
+    description: "Only applicable to FRP. Library-specific assay configuration. By\
+      \ default,\nthe assay configuration is detected automatically. Typically, users\
+      \ will\nnot need to specify a chemistry.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Sample parameters"
+  arguments:
+  - type: "string"
+    name: "--sample_ids"
+    alternatives:
+    - "--cell_multiplex_sample_id"
+    description: "A name to identify a multiplexed sample. Must be alphanumeric with\
+      \ hyphens and/or underscores,\nand less than 64 characters. Required for Cell\
+      \ Multiplexing libraries.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--sample_description"
+    alternatives:
+    - "--cell_multiplex_description"
+    description: "A description for the sample."
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--sample_expect_cells"
+    description: "Expected number of recovered cells, used as input to cell calling\
+      \ algorithm.\n"
+    info: null
+    example:
+    - 3000
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--sample_force_cells"
+    description: "Force pipeline to use this number of cells, bypassing cell detection.\n"
+    info: null
+    example:
+    - 3000
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Feature Barcode library specific arguments"
+  arguments:
+  - type: "file"
+    name: "--feature_reference"
+    description: "Path to the Feature reference CSV file, declaring Feature Barcode\
+      \ constructs and associated barcodes.\nRequired only for Antibody Capture or\
+      \ CRISPR Guide Capture libraries.\nSee https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/using/feature-bc-analysis#feature-ref\
+      \ for more information.\"\n"
+    info: null
+    example:
+    - "feature_reference.csv"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--feature_r1_length"
+    description: "Limit the length of the input Read 1 sequence of V(D)J libraries\
+      \ to the first N bases,\nwhere N is the user-supplied value. Note that the length\
+      \ includes the Barcode and UMI\nsequences so do not set this below 26.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--feature_r2_length"
+    description: "Limit the length of the input Read 2 sequence of V(D)J libraries\
+      \ to the first N bases,\nwhere N is a user-supplied value. Trimming occurs before\
+      \ sequencing metrics are computed\nand therefore, limiting the length of Read\
+      \ 2 may affect Q30 scores.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_crispr_umi"
+    description: "Set the minimum number of CRISPR guide RNA UMIs required for protospacer\
+      \ detection.\nIf a lower or higher sensitivity is desired for detection, this\
+      \ value can be customized\naccording to specific experimental needs. Applicable\
+      \ only to datasets that include a\nCRISPR Guide Capture library.\n"
+    info: null
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Gene expression arguments"
+  description: "Arguments relevant to the analysis of gene expression data."
+  arguments:
+  - type: "file"
+    name: "--gex_reference"
+    description: "Genome refence index built by Cell Ranger mkref."
+    info: null
+    example:
+    - "reference_genome.tar.gz"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--gex_secondary_analysis"
+    description: "Whether or not to run the secondary analysis e.g. clustering."
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--gex_generate_bam"
+    description: "Whether to generate a BAM file."
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--tenx_cloud_token_path"
+    description: "The 10x Cloud Analysis user token used to enable cell annotation."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--cell_annotation_model"
+    description: "\"Cell annotation model to use. If auto, uses the default model\
+      \ for the species.\nIf not given, does not run cell annotation.\"\n"
+    info: null
+    required: false
+    choices:
+    - "auto"
+    - "human_pca_v1_beta"
+    - "mouse_pca_v1_beta"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gex_expect_cells"
+    description: "Expected number of recovered cells, used as input to cell calling\
+      \ algorithm.\n"
+    info: null
+    example:
+    - 3000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gex_force_cells"
+    description: "Force pipeline to use this number of cells, bypassing cell detection.\n"
+    info: null
+    example:
+    - 3000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--gex_include_introns"
+    description: "Whether or not to include intronic reads in counts.\nThis option\
+      \ does not apply to Fixed RNA Profiling analysis.\n"
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gex_r1_length"
+    description: "Limit the length of the input Read 1 sequence of V(D)J libraries\
+      \ to the first N bases,\nwhere N is the user-supplied value. Note that the length\
+      \ includes the Barcode and UMI\nsequences so do not set this below 26.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gex_r2_length"
+    description: "Limit the length of the input Read 2 sequence of V(D)J libraries\
+      \ to the first N bases,\nwhere N is a user-supplied value. Trimming occurs before\
+      \ sequencing metrics are computed\nand therefore, limiting the length of Read\
+      \ 2 may affect Q30 scores.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--gex_chemistry"
+    description: "Assay configuration. Either specify a single value which will be\
+      \ applied to all libraries,\nor a number of values that is equal to the number\
+      \ of libararies. The latter is only applicable\nto only applicable to Fixed\
+      \ RNA Profiling.\n  - auto: Chemistry autodetection (default)\n  - threeprime:\
+      \ Single Cell 3'\n  - SC3Pv1, SC3Pv2, SC3Pv3(-polyA), SC3Pv4(-polyA): Single\
+      \ Cell 3' v1, v2, v3, or v4\n  - SC3Pv3HT(-polyA): Single Cell 3' v3.1 HT\n\
+      \  - SC-FB: Single Cell Antibody-only 3' v2 or 5'\n  - fiveprime: Single Cell\
+      \ 5'\n  - SC5P-PE: Paired-end Single Cell 5'\n  - SC5P-PE-v3: Paired-end Single\
+      \ Cell 5' v3\n  - SC5P-R2: R2-only Single Cell 5'\n  - SC5P-R2-v3: R2-only Single\
+      \ Cell 5' v3\n  - SCP5-PE-v3: Single Cell 5' paired-end v3 (GEM-X)\n  - SC5PHT\
+      \ : Single Cell 5' v2 HT\n  - SFRP: Fixed RNA Profiling (Singleplex)\n  - MFRP:\
+      \ Fixed RNA Profiling (Multiplex, Probe Barcode on R2)\n  - MFRP-R1: Fixed RNA\
+      \ Profiling (Multiplex, Probe Barcode on R1)\n  - MFRP-RNA: Fixed RNA Profiling\
+      \ (Multiplex, RNA, Probe Barcode on R2)\n  - MFRP-Ab: Fixed RNA Profiling (Multiplex,\
+      \ Antibody, Probe Barcode at R2:69)\n  - MFRP-Ab-R2pos50: Fixed RNA Profiling\
+      \ (Multiplex, Antibody, Probe Barcode at R2:50)\n  - MFRP-RNA-R1: Fixed RNA\
+      \ Profiling (Multiplex, RNA, Probe Barcode on R1)\n  - MFRP-Ab-R1: Fixed RNA\
+      \ Profiling (Multiplex, Antibody, Probe Barcode on R1)\n  - ARC-v1 for analyzing\
+      \ the Gene Expression portion of Multiome data. If Cell Ranger auto-detects\
+      \ ARC-v1 chemistry, an error is triggered.\nSee https://kb.10xgenomics.com/hc/en-us/articles/115003764132-How-does-Cell-Ranger-auto-detect-chemistry-\
+      \ for more information.\n"
+    info: null
+    default:
+    - "auto"
+    required: false
+    choices:
+    - "auto"
+    - "threeprime"
+    - "fiveprime"
+    - "SC3Pv1"
+    - "SC3Pv2"
+    - "SC3Pv3"
+    - "SC3Pv3-polyA"
+    - "SC3Pv4"
+    - "SC3Pv4-polyA"
+    - "SC3Pv3LT"
+    - "SC3Pv3HT"
+    - "SC3Pv3HT-polyA"
+    - "SC5P-PE"
+    - "SC5P-PE-v3"
+    - "SC5P-R2"
+    - "SC-FB"
+    - "SC5P-R2-v3"
+    - "SCP5-PE-v3"
+    - "SC5PHT"
+    - "MFRP"
+    - "MFRP-R1"
+    - "MFRP-RNA"
+    - "MFRP-Ab"
+    - "SFRP"
+    - "MFRP-Ab-R2pos50"
+    - "MFRP-RNA-R1"
+    - "MFRP-Ab-R1"
+    - "ARC-v1"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "VDJ related parameters"
+  arguments:
+  - type: "file"
+    name: "--vdj_reference"
+    description: "VDJ refence index built by Cell Ranger mkref."
+    info: null
+    example:
+    - "reference_vdj.tar.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--vdj_inner_enrichment_primers"
+    description: "V(D)J Immune Profiling libraries: if inner enrichment primers other\
+      \ than those provided \nin the 10x Genomics kits are used, they need to be specified\
+      \ here as a\ntext file with one primer per line.\n"
+    info: null
+    example:
+    - "enrichment_primers.txt"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--vdj_r1_length"
+    description: "Limit the length of the input Read 1 sequence of V(D)J libraries\
+      \ to the first N bases, where N is the user-supplied value.\nNote that the length\
+      \ includes the Barcode and UMI sequences so do not set this below 26.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--vdj_r2_length"
+    description: "Limit the length of the input Read 2 sequence of V(D)J libraries\
+      \ to the first N bases, where N is a user-supplied value. \nTrimming occurs\
+      \ before sequencing metrics are computed and therefore, limiting the length\
+      \ of Read 2 may affect Q30 scores\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--vdj_denovo"
+    description: "Run in reference-free mode (i.e., do not use annotations). This\
+      \ option is not supported for multiplexed experiments.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "3' Cell multiplexing parameters (CellPlex Multiplexing)"
+  arguments:
+  - type: "string"
+    name: "--cell_multiplex_oligo_ids"
+    alternatives:
+    - "--cmo_ids"
+    description: "The Cell Multiplexing oligo IDs used to multiplex this sample. If\
+      \ multiple CMOs were used for a sample,\nseparate IDs with a pipe (e.g., CMO301|CMO302).\
+      \ Required for Cell Multiplexing libraries.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "double"
+    name: "--min_assignment_confidence"
+    description: "The minimum estimated likelihood to call a sample as tagged with\
+      \ a Cell Multiplexing Oligo (CMO) instead of \"Unassigned\".\nUsers may wish\
+      \ to tolerate a higher rate of mis-assignment in order to obtain more singlets\
+      \ to include in their analysis,\nor a lower rate of mis-assignment at the cost\
+      \ of obtaining fewer singlets.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--cmo_set"
+    description: "Path to a custom CMO set CSV file, declaring CMO constructs and\
+      \ associated barcodes. If the default CMO reference IDs that are built into\n\
+      the Cell Ranger software are required, this option does not need to be used.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--barcode_sample_assignment"
+    description: "Path to a barcode-sample assignment CSV file that specifies the\
+      \ barcodes that belong to each sample.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Hashtag multiplexing parameters"
+  arguments:
+  - type: "string"
+    name: "--hashtag_ids"
+    description: "The hashtag IDs used to multiplex this sample. If multiple antibody\
+      \ hashtags were used for the same sample,\nyou can separate IDs with a pipe.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "On-chip multiplexing parameters"
+  arguments:
+  - type: "string"
+    name: "--ocm_barcode_ids"
+    description: "The OCM barcode IDs used to multiplex this sample. Must be one of\
+      \ OB1, OB2, OB3, OB4.\nIf multiple OCM Barcodes were used for the same sample,\
+      \ you can separate IDs\nwith a pipe (e.g., OB1|OB2).\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Flex multiplexing paramaters"
+  arguments:
+  - type: "file"
+    name: "--probe_set"
+    description: "A probe set reference CSV file. It specifies the sequences used\
+      \ as a reference for probe alignment and the gene ID associated with each probe.\n\
+      It must include 4 columns (probe file format 1.0.0): gene_id,probe_seq,probe_id,included,region\
+      \ and an optional 5th column (probe file format 1.0.1).\n- gene_id: The Ensembl\
+      \ gene identifier targeted by the probe.\n- probe_seq: The nucleotide sequence\
+      \ of the probe, which is complementary to the transcript sequence.\n- probe_id:\
+      \ The probe identifier, whose format is described in Probe identifiers.\n- included:\
+      \ A TRUE or FALSE flag specifying whether the probe is included in the filtered\
+      \ counts matrix output or excluded by the probe filter. \n            See filter-probes\
+      \ option of cellranger multi. All probes of a gene must be marked TRUE in the\
+      \ included column for that gene to be included.\n- region: Present only in v1.0.1\
+      \ probe set reference CSV. The gene boundary targeted by the probe. Accepted\
+      \ values are spliced or unspliced.\n\nThe file also contains a number of required\
+      \ metadata fields in the header in the format #key=value:\n- panel_name: The\
+      \ name of the probe set.\n- panel_type: Always predesigned for predesigned probe\
+      \ sets.\n- reference_genome: The reference genome build used for probe design.\n\
+      - reference_version: The version of the Cell Ranger reference transcriptome\
+      \ used for probe design.\n- probe_set_file_format: The version of the probe\
+      \ set file format specification that this file conforms to.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--filter_probes"
+    description: "If 'false', include all non-deprecated probes listed in the probe\
+      \ set reference CSV file.\nIf 'true' or not set, probes that are predicted to\
+      \ have off-target activity to homologous genes are excluded from analysis.\n\
+      Not filtering will result in UMI counts from all non-deprecated probes,\nincluding\
+      \ those with predicted off-target activity, to be used in the analysis.\nProbes\
+      \ whose ID is prefixed with DEPRECATED are always excluded from the analysis.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--probe_barcode_ids"
+    description: "The Fixed RNA Probe Barcode ID used for this sample, and for multiplex\
+      \ GEX + Antibody Capture libraries,\nthe corresponding Antibody Multiplexing\
+      \ Barcode IDs. 10x recommends specifying both barcodes (e.g., BC001+AB001)\n\
+      when an Antibody Capture library is present. The barcode pair order is BC+AB\
+      \ and they\nare separated with a \"+\" (no spaces). Alternatively, you can specify\
+      \ the Probe Barcode ID alone and\nCell Ranger's barcode pairing auto-detection\
+      \ algorithm will automatically match to the corresponding Antibody\nMultiplexing\
+      \ Barcode.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--emptydrops_minimum_umis"
+    description: "For singleplex Flex experiments, use this option to adjust the UMI\
+      \ cutoff during the second step of cell calling.\nCell Ranger will still perform\
+      \ the full cell calling process but will only evaluate barcodes with UMIs above\n\
+      the threshold you specify.\n"
+    info: null
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Antigen Capture (BEAM) libary arguments"
+  description: "These arguments are recommended if an Antigen Capture (BEAM) library\
+    \ is present. \nIt is needed to calculate the antigen specificity score.\n"
+  arguments:
+  - type: "string"
+    name: "--control_id"
+    description: "A user-defined ID for any negative controls used in the T/BCR Antigen\
+      \ Capture assay. Must match id specified in the feature reference CSV.\nMay\
+      \ only include ASCII characters and must not use whitespace, slash, quote, or\
+      \ comma characters. \nEach ID must be unique and must not collide with a gene\
+      \ identifier from the transcriptome.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--mhc_allele"
+    description: "The MHC allele for TCR Antigen Capture libraries. Must match mhc_allele\
+      \ name specified in the Feature Reference CSV.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "General arguments"
+  description: "These arguments are applicable to all library types.\n"
+  arguments:
+  - type: "boolean"
+    name: "--check_library_compatibility"
+    description: "Optional. This option allows users to disable the check that evaluates\
+      \ 10x Barcode overlap between\nibraries when multiple libraries are specified\
+      \ (e.g., Gene Expression + Antibody Capture). Setting\nthis option to false\
+      \ will disable the check across all library combinations. We recommend running\n\
+      this check (default), however if the pipeline errors out, users can bypass the\
+      \ check to generate\noutputs for troubleshooting.\n"
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "The folder to store the alignment results."
+    info: null
+    example:
+    - "/path/to/output"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Executor arguments"
+  arguments:
+  - type: "boolean_true"
+    name: "--dryrun"
+    description: "If true, the output directory will only contain the CWL input files,\
+      \ but the pipeline itself will not be executed."
+    info: null
+    direction: "input"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Align fastq files using Cell Ranger multi."
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "10x_5k_anticmv/raw/"
+  dest: "10x_5k_anticmv/raw/"
+- type: "file"
+  path: "10x_5k_lung_crispr/raw/"
+  dest: "10x_5k_lung_crispr/raw/"
+- type: "file"
+  path: "10x_5k_beam/raw/"
+  dest: "10x_5k_beam/raw/"
+- type: "file"
+  path: "10x_5k_fixed/raw"
+  dest: "10x_5k_fixed/raw"
+- type: "file"
+  path: "raw"
+- type: "file"
+  path: "reference_gencodev41_chr1"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "veryhighmem"
+    - "highcpu"
+    - "highdisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "quay.io/nf-core/cellranger:10.0.0"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "docker"
+    run:
+    - "cellranger telemetry disable"
+  - type: "python"
+    user: false
+    packages:
+    - "pandas"
+    - "pyyaml"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/mapping/cellranger_multi/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/mapping/cellranger_multi"
+  executable: "target/nextflow/mapping/cellranger_multi/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/main.nf
new file mode 100644
index 0000000..ab48dfc
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/main.nf
@@ -0,0 +1,5273 @@
+// cellranger_multi v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+//  * Angela Oliveira Pisco (author)
+//  * Robrecht Cannoodt (author, maintainer)
+//  * Dries De Maeyer (author)
+//  * Weiwei Schultz (contributor)
+//  * Dorien Roosen (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "cellranger_multi",
+  "namespace" : "mapping",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Angela Oliveira Pisco",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "links" : {
+          "github" : "aopisco",
+          "orcid" : "0000-0003-0142-2355",
+          "linkedin" : "aopisco"
+        },
+        "organizations" : [
+          {
+            "name" : "Insitro",
+            "href" : "https://insitro.com",
+            "role" : "Director of Computational Biology"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Input files",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "The FASTQ files to be analyzed. FASTQ files should conform to the naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Feature type-specific input files",
+      "description" : "Helper functionality to allow feature type-specific input files, without the need to specify\nlibrary_type or library_id. The library_id will be inferred from the input paths.\n",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--gex_input",
+          "description" : "The FASTQ files to be analyzed for Gene Expression. FASTQ files should conform to the\nnaming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--abc_input",
+          "description" : "The FASTQ files to be analyzed for Antibody Capture. FASTQ files should conform to \nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--cgc_input",
+          "description" : "The FASTQ files to be analyzed for CRISPR Guide Capture. FASTQ files should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--mux_input",
+          "description" : "The FASTQ files to be analyzed for Multiplexing Capture. FASTQ files should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--vdj_input",
+          "description" : "The FASTQ files to be analyzed for VDJ. FASTQ files should conform to the\nnaming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--vdj_t_input",
+          "description" : "The FASTQ files to be analyzed for VDJ-T. FASTQ files should conform to the naming\nconventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--vdj_t_gd_input",
+          "description" : "The FASTQ files to be analyzed for VDJ-T-GD. FASTQ files should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--vdj_b_input",
+          "description" : "The FASTQ files to be analyzed for VDJ-B. FASTQ files should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--agc_input",
+          "description" : "The FASTQ files to be analyzed for Antigen Capture. FASTQ files should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Library arguments",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--library_id",
+          "description" : "The Illumina sample name to analyze. This must exactly match the 'Sample Name'part\nof the FASTQ files specified in the `--input` argument.\n",
+          "example" : [
+            "mysample1"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--library_type",
+          "description" : "The underlying feature type of the library.\n",
+          "example" : [
+            "Gene Expression"
+          ],
+          "required" : false,
+          "choices" : [
+            "Gene Expression",
+            "VDJ",
+            "VDJ-T",
+            "VDJ-B",
+            "VDJ-T-GD",
+            "Antibody Capture",
+            "CRISPR Guide Capture",
+            "Multiplexing Capture",
+            "Antigen Capture",
+            "Custom"
+          ],
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--library_subsample",
+          "description" : "The rate at which reads from the provided FASTQ files are sampled.\nMust be strictly greater than 0 and less than or equal to 1.\n",
+          "example" : [
+            "0.5"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--library_lanes",
+          "description" : "Lanes associated with this sample. Defaults to using all lanes.",
+          "example" : [
+            "1-4"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--library_chemistry",
+          "description" : "Only applicable to FRP. Library-specific assay configuration. By default,\nthe assay configuration is detected automatically. Typically, users will\nnot need to specify a chemistry.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Sample parameters",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--sample_ids",
+          "alternatives" : [
+            "--cell_multiplex_sample_id"
+          ],
+          "description" : "A name to identify a multiplexed sample. Must be alphanumeric with hyphens and/or underscores,\nand less than 64 characters. Required for Cell Multiplexing libraries.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--sample_description",
+          "alternatives" : [
+            "--cell_multiplex_description"
+          ],
+          "description" : "A description for the sample.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--sample_expect_cells",
+          "description" : "Expected number of recovered cells, used as input to cell calling algorithm.\n",
+          "example" : [
+            3000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--sample_force_cells",
+          "description" : "Force pipeline to use this number of cells, bypassing cell detection.\n",
+          "example" : [
+            3000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Feature Barcode library specific arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--feature_reference",
+          "description" : "Path to the Feature reference CSV file, declaring Feature Barcode constructs and associated barcodes.\nRequired only for Antibody Capture or CRISPR Guide Capture libraries.\nSee https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/using/feature-bc-analysis#feature-ref for more information.\\"\n",
+          "example" : [
+            "feature_reference.csv"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--feature_r1_length",
+          "description" : "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,\nwhere N is the user-supplied value. Note that the length includes the Barcode and UMI\nsequences so do not set this below 26.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--feature_r2_length",
+          "description" : "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,\nwhere N is a user-supplied value. Trimming occurs before sequencing metrics are computed\nand therefore, limiting the length of Read 2 may affect Q30 scores.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_crispr_umi",
+          "description" : "Set the minimum number of CRISPR guide RNA UMIs required for protospacer detection.\nIf a lower or higher sensitivity is desired for detection, this value can be customized\naccording to specific experimental needs. Applicable only to datasets that include a\nCRISPR Guide Capture library.\n",
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Gene expression arguments",
+      "description" : "Arguments relevant to the analysis of gene expression data.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--gex_reference",
+          "description" : "Genome refence index built by Cell Ranger mkref.",
+          "example" : [
+            "reference_genome.tar.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--gex_secondary_analysis",
+          "description" : "Whether or not to run the secondary analysis e.g. clustering.",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--gex_generate_bam",
+          "description" : "Whether to generate a BAM file.",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--tenx_cloud_token_path",
+          "description" : "The 10x Cloud Analysis user token used to enable cell annotation.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--cell_annotation_model",
+          "description" : "\\"Cell annotation model to use. If auto, uses the default model for the species.\nIf not given, does not run cell annotation.\\"\n",
+          "required" : false,
+          "choices" : [
+            "auto",
+            "human_pca_v1_beta",
+            "mouse_pca_v1_beta"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gex_expect_cells",
+          "description" : "Expected number of recovered cells, used as input to cell calling algorithm.\n",
+          "example" : [
+            3000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gex_force_cells",
+          "description" : "Force pipeline to use this number of cells, bypassing cell detection.\n",
+          "example" : [
+            3000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--gex_include_introns",
+          "description" : "Whether or not to include intronic reads in counts.\nThis option does not apply to Fixed RNA Profiling analysis.\n",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gex_r1_length",
+          "description" : "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,\nwhere N is the user-supplied value. Note that the length includes the Barcode and UMI\nsequences so do not set this below 26.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gex_r2_length",
+          "description" : "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,\nwhere N is a user-supplied value. Trimming occurs before sequencing metrics are computed\nand therefore, limiting the length of Read 2 may affect Q30 scores.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--gex_chemistry",
+          "description" : "Assay configuration. Either specify a single value which will be applied to all libraries,\nor a number of values that is equal to the number of libararies. The latter is only applicable\nto only applicable to Fixed RNA Profiling.\n  - auto: Chemistry autodetection (default)\n  - threeprime: Single Cell 3'\n  - SC3Pv1, SC3Pv2, SC3Pv3(-polyA), SC3Pv4(-polyA): Single Cell 3' v1, v2, v3, or v4\n  - SC3Pv3HT(-polyA): Single Cell 3' v3.1 HT\n  - SC-FB: Single Cell Antibody-only 3' v2 or 5'\n  - fiveprime: Single Cell 5'\n  - SC5P-PE: Paired-end Single Cell 5'\n  - SC5P-PE-v3: Paired-end Single Cell 5' v3\n  - SC5P-R2: R2-only Single Cell 5'\n  - SC5P-R2-v3: R2-only Single Cell 5' v3\n  - SCP5-PE-v3: Single Cell 5' paired-end v3 (GEM-X)\n  - SC5PHT : Single Cell 5' v2 HT\n  - SFRP: Fixed RNA Profiling (Singleplex)\n  - MFRP: Fixed RNA Profiling (Multiplex, Probe Barcode on R2)\n  - MFRP-R1: Fixed RNA Profiling (Multiplex, Probe Barcode on R1)\n  - MFRP-RNA: Fixed RNA Profiling (Multiplex, RNA, Probe Barcode on R2)\n  - MFRP-Ab: Fixed RNA Profiling (Multiplex, Antibody, Probe Barcode at R2:69)\n  - MFRP-Ab-R2pos50: Fixed RNA Profiling (Multiplex, Antibody, Probe Barcode at R2:50)\n  - MFRP-RNA-R1: Fixed RNA Profiling (Multiplex, RNA, Probe Barcode on R1)\n  - MFRP-Ab-R1: Fixed RNA Profiling (Multiplex, Antibody, Probe Barcode on R1)\n  - ARC-v1 for analyzing the Gene Expression portion of Multiome data. If Cell Ranger auto-detects ARC-v1 chemistry, an error is triggered.\nSee https://kb.10xgenomics.com/hc/en-us/articles/115003764132-How-does-Cell-Ranger-auto-detect-chemistry- for more information.\n",
+          "default" : [
+            "auto"
+          ],
+          "required" : false,
+          "choices" : [
+            "auto",
+            "threeprime",
+            "fiveprime",
+            "SC3Pv1",
+            "SC3Pv2",
+            "SC3Pv3",
+            "SC3Pv3-polyA",
+            "SC3Pv4",
+            "SC3Pv4-polyA",
+            "SC3Pv3LT",
+            "SC3Pv3HT",
+            "SC3Pv3HT-polyA",
+            "SC5P-PE",
+            "SC5P-PE-v3",
+            "SC5P-R2",
+            "SC-FB",
+            "SC5P-R2-v3",
+            "SCP5-PE-v3",
+            "SC5PHT",
+            "MFRP",
+            "MFRP-R1",
+            "MFRP-RNA",
+            "MFRP-Ab",
+            "SFRP",
+            "MFRP-Ab-R2pos50",
+            "MFRP-RNA-R1",
+            "MFRP-Ab-R1",
+            "ARC-v1"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "VDJ related parameters",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--vdj_reference",
+          "description" : "VDJ refence index built by Cell Ranger mkref.",
+          "example" : [
+            "reference_vdj.tar.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--vdj_inner_enrichment_primers",
+          "description" : "V(D)J Immune Profiling libraries: if inner enrichment primers other than those provided \nin the 10x Genomics kits are used, they need to be specified here as a\ntext file with one primer per line.\n",
+          "example" : [
+            "enrichment_primers.txt"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--vdj_r1_length",
+          "description" : "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases, where N is the user-supplied value.\nNote that the length includes the Barcode and UMI sequences so do not set this below 26.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--vdj_r2_length",
+          "description" : "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases, where N is a user-supplied value. \nTrimming occurs before sequencing metrics are computed and therefore, limiting the length of Read 2 may affect Q30 scores\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--vdj_denovo",
+          "description" : "Run in reference-free mode (i.e., do not use annotations). This option is not supported for multiplexed experiments.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "3' Cell multiplexing parameters (CellPlex Multiplexing)",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--cell_multiplex_oligo_ids",
+          "alternatives" : [
+            "--cmo_ids"
+          ],
+          "description" : "The Cell Multiplexing oligo IDs used to multiplex this sample. If multiple CMOs were used for a sample,\nseparate IDs with a pipe (e.g., CMO301|CMO302). Required for Cell Multiplexing libraries.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--min_assignment_confidence",
+          "description" : "The minimum estimated likelihood to call a sample as tagged with a Cell Multiplexing Oligo (CMO) instead of \\"Unassigned\\".\nUsers may wish to tolerate a higher rate of mis-assignment in order to obtain more singlets to include in their analysis,\nor a lower rate of mis-assignment at the cost of obtaining fewer singlets.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--cmo_set",
+          "description" : "Path to a custom CMO set CSV file, declaring CMO constructs and associated barcodes. If the default CMO reference IDs that are built into\nthe Cell Ranger software are required, this option does not need to be used.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--barcode_sample_assignment",
+          "description" : "Path to a barcode-sample assignment CSV file that specifies the barcodes that belong to each sample.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Hashtag multiplexing parameters",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--hashtag_ids",
+          "description" : "The hashtag IDs used to multiplex this sample. If multiple antibody hashtags were used for the same sample,\nyou can separate IDs with a pipe.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "On-chip multiplexing parameters",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--ocm_barcode_ids",
+          "description" : "The OCM barcode IDs used to multiplex this sample. Must be one of OB1, OB2, OB3, OB4.\nIf multiple OCM Barcodes were used for the same sample, you can separate IDs\nwith a pipe (e.g., OB1|OB2).\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Flex multiplexing paramaters",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--probe_set",
+          "description" : "A probe set reference CSV file. It specifies the sequences used as a reference for probe alignment and the gene ID associated with each probe.\nIt must include 4 columns (probe file format 1.0.0): gene_id,probe_seq,probe_id,included,region and an optional 5th column (probe file format 1.0.1).\n- gene_id: The Ensembl gene identifier targeted by the probe.\n- probe_seq: The nucleotide sequence of the probe, which is complementary to the transcript sequence.\n- probe_id: The probe identifier, whose format is described in Probe identifiers.\n- included: A TRUE or FALSE flag specifying whether the probe is included in the filtered counts matrix output or excluded by the probe filter. \n            See filter-probes option of cellranger multi. All probes of a gene must be marked TRUE in the included column for that gene to be included.\n- region: Present only in v1.0.1 probe set reference CSV. The gene boundary targeted by the probe. Accepted values are spliced or unspliced.\n\nThe file also contains a number of required metadata fields in the header in the format #key=value:\n- panel_name: The name of the probe set.\n- panel_type: Always predesigned for predesigned probe sets.\n- reference_genome: The reference genome build used for probe design.\n- reference_version: The version of the Cell Ranger reference transcriptome used for probe design.\n- probe_set_file_format: The version of the probe set file format specification that this file conforms to.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--filter_probes",
+          "description" : "If 'false', include all non-deprecated probes listed in the probe set reference CSV file.\nIf 'true' or not set, probes that are predicted to have off-target activity to homologous genes are excluded from analysis.\nNot filtering will result in UMI counts from all non-deprecated probes,\nincluding those with predicted off-target activity, to be used in the analysis.\nProbes whose ID is prefixed with DEPRECATED are always excluded from the analysis.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--probe_barcode_ids",
+          "description" : "The Fixed RNA Probe Barcode ID used for this sample, and for multiplex GEX + Antibody Capture libraries,\nthe corresponding Antibody Multiplexing Barcode IDs. 10x recommends specifying both barcodes (e.g., BC001+AB001)\nwhen an Antibody Capture library is present. The barcode pair order is BC+AB and they\nare separated with a \\"+\\" (no spaces). Alternatively, you can specify the Probe Barcode ID alone and\nCell Ranger's barcode pairing auto-detection algorithm will automatically match to the corresponding Antibody\nMultiplexing Barcode.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--emptydrops_minimum_umis",
+          "description" : "For singleplex Flex experiments, use this option to adjust the UMI cutoff during the second step of cell calling.\nCell Ranger will still perform the full cell calling process but will only evaluate barcodes with UMIs above\nthe threshold you specify.\n",
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Antigen Capture (BEAM) libary arguments",
+      "description" : "These arguments are recommended if an Antigen Capture (BEAM) library is present. \nIt is needed to calculate the antigen specificity score.\n",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--control_id",
+          "description" : "A user-defined ID for any negative controls used in the T/BCR Antigen Capture assay. Must match id specified in the feature reference CSV.\nMay only include ASCII characters and must not use whitespace, slash, quote, or comma characters. \nEach ID must be unique and must not collide with a gene identifier from the transcriptome.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--mhc_allele",
+          "description" : "The MHC allele for TCR Antigen Capture libraries. Must match mhc_allele name specified in the Feature Reference CSV.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "General arguments",
+      "description" : "These arguments are applicable to all library types.\n",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--check_library_compatibility",
+          "description" : "Optional. This option allows users to disable the check that evaluates 10x Barcode overlap between\nibraries when multiple libraries are specified (e.g., Gene Expression + Antibody Capture). Setting\nthis option to false will disable the check across all library combinations. We recommend running\nthis check (default), however if the pipeline errors out, users can bypass the check to generate\noutputs for troubleshooting.\n",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "The folder to store the alignment results.",
+          "example" : [
+            "/path/to/output"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Executor arguments",
+      "arguments" : [
+        {
+          "type" : "boolean_true",
+          "name" : "--dryrun",
+          "description" : "If true, the output directory will only contain the CWL input files, but the pipeline itself will not be executed.",
+          "direction" : "input"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Align fastq files using Cell Ranger multi.",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_anticmv/raw/",
+      "dest" : "10x_5k_anticmv/raw/"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_lung_crispr/raw/",
+      "dest" : "10x_5k_lung_crispr/raw/"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_beam/raw/",
+      "dest" : "10x_5k_beam/raw/"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_fixed/raw/",
+      "dest" : "10x_5k_fixed/raw"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_beam/raw/"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/reference_gencodev41_chr1"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "veryhighmem",
+          "highcpu",
+          "highdisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "quay.io/nf-core/cellranger:10.0.0",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "docker",
+          "run" : [
+            "cellranger telemetry disable"
+          ]
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "pandas",
+            "pyyaml"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/mapping/cellranger_multi/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/mapping/cellranger_multi",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+from __future__ import annotations
+
+import sys
+import re
+import subprocess
+from multiprocessing import cpu_count
+import tempfile
+import pandas as pd
+import yaml
+from typing import Optional, Any, Union
+import tarfile
+from pathlib import Path
+import shutil
+from itertools import chain
+
+CELL_RANGER_EXEC = shutil.which("cellranger")
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'gex_input': $( if [ ! -z ${VIASH_PAR_GEX_INPUT+x} ]; then echo "r'${VIASH_PAR_GEX_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'abc_input': $( if [ ! -z ${VIASH_PAR_ABC_INPUT+x} ]; then echo "r'${VIASH_PAR_ABC_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'cgc_input': $( if [ ! -z ${VIASH_PAR_CGC_INPUT+x} ]; then echo "r'${VIASH_PAR_CGC_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'mux_input': $( if [ ! -z ${VIASH_PAR_MUX_INPUT+x} ]; then echo "r'${VIASH_PAR_MUX_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'vdj_input': $( if [ ! -z ${VIASH_PAR_VDJ_INPUT+x} ]; then echo "r'${VIASH_PAR_VDJ_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'vdj_t_input': $( if [ ! -z ${VIASH_PAR_VDJ_T_INPUT+x} ]; then echo "r'${VIASH_PAR_VDJ_T_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'vdj_t_gd_input': $( if [ ! -z ${VIASH_PAR_VDJ_T_GD_INPUT+x} ]; then echo "r'${VIASH_PAR_VDJ_T_GD_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'vdj_b_input': $( if [ ! -z ${VIASH_PAR_VDJ_B_INPUT+x} ]; then echo "r'${VIASH_PAR_VDJ_B_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'agc_input': $( if [ ! -z ${VIASH_PAR_AGC_INPUT+x} ]; then echo "r'${VIASH_PAR_AGC_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'library_id': $( if [ ! -z ${VIASH_PAR_LIBRARY_ID+x} ]; then echo "r'${VIASH_PAR_LIBRARY_ID//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'library_type': $( if [ ! -z ${VIASH_PAR_LIBRARY_TYPE+x} ]; then echo "r'${VIASH_PAR_LIBRARY_TYPE//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'library_subsample': $( if [ ! -z ${VIASH_PAR_LIBRARY_SUBSAMPLE+x} ]; then echo "r'${VIASH_PAR_LIBRARY_SUBSAMPLE//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'library_lanes': $( if [ ! -z ${VIASH_PAR_LIBRARY_LANES+x} ]; then echo "r'${VIASH_PAR_LIBRARY_LANES//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'library_chemistry': $( if [ ! -z ${VIASH_PAR_LIBRARY_CHEMISTRY+x} ]; then echo "r'${VIASH_PAR_LIBRARY_CHEMISTRY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'sample_ids': $( if [ ! -z ${VIASH_PAR_SAMPLE_IDS+x} ]; then echo "r'${VIASH_PAR_SAMPLE_IDS//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'sample_description': $( if [ ! -z ${VIASH_PAR_SAMPLE_DESCRIPTION+x} ]; then echo "r'${VIASH_PAR_SAMPLE_DESCRIPTION//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'sample_expect_cells': $( if [ ! -z ${VIASH_PAR_SAMPLE_EXPECT_CELLS+x} ]; then echo "list(map(int, r'${VIASH_PAR_SAMPLE_EXPECT_CELLS//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'sample_force_cells': $( if [ ! -z ${VIASH_PAR_SAMPLE_FORCE_CELLS+x} ]; then echo "list(map(int, r'${VIASH_PAR_SAMPLE_FORCE_CELLS//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'feature_reference': $( if [ ! -z ${VIASH_PAR_FEATURE_REFERENCE+x} ]; then echo "r'${VIASH_PAR_FEATURE_REFERENCE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'feature_r1_length': $( if [ ! -z ${VIASH_PAR_FEATURE_R1_LENGTH+x} ]; then echo "int(r'${VIASH_PAR_FEATURE_R1_LENGTH//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'feature_r2_length': $( if [ ! -z ${VIASH_PAR_FEATURE_R2_LENGTH+x} ]; then echo "int(r'${VIASH_PAR_FEATURE_R2_LENGTH//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'min_crispr_umi': $( if [ ! -z ${VIASH_PAR_MIN_CRISPR_UMI+x} ]; then echo "int(r'${VIASH_PAR_MIN_CRISPR_UMI//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'gex_reference': $( if [ ! -z ${VIASH_PAR_GEX_REFERENCE+x} ]; then echo "r'${VIASH_PAR_GEX_REFERENCE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'gex_secondary_analysis': $( if [ ! -z ${VIASH_PAR_GEX_SECONDARY_ANALYSIS+x} ]; then echo "r'${VIASH_PAR_GEX_SECONDARY_ANALYSIS//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'gex_generate_bam': $( if [ ! -z ${VIASH_PAR_GEX_GENERATE_BAM+x} ]; then echo "r'${VIASH_PAR_GEX_GENERATE_BAM//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'tenx_cloud_token_path': $( if [ ! -z ${VIASH_PAR_TENX_CLOUD_TOKEN_PATH+x} ]; then echo "r'${VIASH_PAR_TENX_CLOUD_TOKEN_PATH//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cell_annotation_model': $( if [ ! -z ${VIASH_PAR_CELL_ANNOTATION_MODEL+x} ]; then echo "r'${VIASH_PAR_CELL_ANNOTATION_MODEL//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'gex_expect_cells': $( if [ ! -z ${VIASH_PAR_GEX_EXPECT_CELLS+x} ]; then echo "int(r'${VIASH_PAR_GEX_EXPECT_CELLS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'gex_force_cells': $( if [ ! -z ${VIASH_PAR_GEX_FORCE_CELLS+x} ]; then echo "int(r'${VIASH_PAR_GEX_FORCE_CELLS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'gex_include_introns': $( if [ ! -z ${VIASH_PAR_GEX_INCLUDE_INTRONS+x} ]; then echo "r'${VIASH_PAR_GEX_INCLUDE_INTRONS//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'gex_r1_length': $( if [ ! -z ${VIASH_PAR_GEX_R1_LENGTH+x} ]; then echo "int(r'${VIASH_PAR_GEX_R1_LENGTH//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'gex_r2_length': $( if [ ! -z ${VIASH_PAR_GEX_R2_LENGTH+x} ]; then echo "int(r'${VIASH_PAR_GEX_R2_LENGTH//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'gex_chemistry': $( if [ ! -z ${VIASH_PAR_GEX_CHEMISTRY+x} ]; then echo "r'${VIASH_PAR_GEX_CHEMISTRY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'vdj_reference': $( if [ ! -z ${VIASH_PAR_VDJ_REFERENCE+x} ]; then echo "r'${VIASH_PAR_VDJ_REFERENCE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'vdj_inner_enrichment_primers': $( if [ ! -z ${VIASH_PAR_VDJ_INNER_ENRICHMENT_PRIMERS+x} ]; then echo "r'${VIASH_PAR_VDJ_INNER_ENRICHMENT_PRIMERS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'vdj_r1_length': $( if [ ! -z ${VIASH_PAR_VDJ_R1_LENGTH+x} ]; then echo "int(r'${VIASH_PAR_VDJ_R1_LENGTH//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'vdj_r2_length': $( if [ ! -z ${VIASH_PAR_VDJ_R2_LENGTH+x} ]; then echo "int(r'${VIASH_PAR_VDJ_R2_LENGTH//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'vdj_denovo': $( if [ ! -z ${VIASH_PAR_VDJ_DENOVO+x} ]; then echo "r'${VIASH_PAR_VDJ_DENOVO//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'cell_multiplex_oligo_ids': $( if [ ! -z ${VIASH_PAR_CELL_MULTIPLEX_OLIGO_IDS+x} ]; then echo "r'${VIASH_PAR_CELL_MULTIPLEX_OLIGO_IDS//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'min_assignment_confidence': $( if [ ! -z ${VIASH_PAR_MIN_ASSIGNMENT_CONFIDENCE+x} ]; then echo "float(r'${VIASH_PAR_MIN_ASSIGNMENT_CONFIDENCE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'cmo_set': $( if [ ! -z ${VIASH_PAR_CMO_SET+x} ]; then echo "r'${VIASH_PAR_CMO_SET//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'barcode_sample_assignment': $( if [ ! -z ${VIASH_PAR_BARCODE_SAMPLE_ASSIGNMENT+x} ]; then echo "r'${VIASH_PAR_BARCODE_SAMPLE_ASSIGNMENT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'hashtag_ids': $( if [ ! -z ${VIASH_PAR_HASHTAG_IDS+x} ]; then echo "r'${VIASH_PAR_HASHTAG_IDS//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'ocm_barcode_ids': $( if [ ! -z ${VIASH_PAR_OCM_BARCODE_IDS+x} ]; then echo "r'${VIASH_PAR_OCM_BARCODE_IDS//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'probe_set': $( if [ ! -z ${VIASH_PAR_PROBE_SET+x} ]; then echo "r'${VIASH_PAR_PROBE_SET//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'filter_probes': $( if [ ! -z ${VIASH_PAR_FILTER_PROBES+x} ]; then echo "r'${VIASH_PAR_FILTER_PROBES//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'probe_barcode_ids': $( if [ ! -z ${VIASH_PAR_PROBE_BARCODE_IDS+x} ]; then echo "r'${VIASH_PAR_PROBE_BARCODE_IDS//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'emptydrops_minimum_umis': $( if [ ! -z ${VIASH_PAR_EMPTYDROPS_MINIMUM_UMIS+x} ]; then echo "int(r'${VIASH_PAR_EMPTYDROPS_MINIMUM_UMIS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'control_id': $( if [ ! -z ${VIASH_PAR_CONTROL_ID+x} ]; then echo "r'${VIASH_PAR_CONTROL_ID//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'mhc_allele': $( if [ ! -z ${VIASH_PAR_MHC_ALLELE+x} ]; then echo "r'${VIASH_PAR_MHC_ALLELE//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'check_library_compatibility': $( if [ ! -z ${VIASH_PAR_CHECK_LIBRARY_COMPATIBILITY+x} ]; then echo "r'${VIASH_PAR_CHECK_LIBRARY_COMPATIBILITY//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'dryrun': $( if [ ! -z ${VIASH_PAR_DRYRUN+x} ]; then echo "r'${VIASH_PAR_DRYRUN//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+if not CELL_RANGER_EXEC:
+    raise RuntimeError(
+        "Could not find Cell Ranger executable. Please make sure it is installed and available in your PATH."
+    )
+
+# Tested with cellranger 7.0:
+# - omitting the lane number is allowed (e.g. \\`_L001\\`)
+# - lane number should be omitted across all files if omitted in one
+# - replacing \\`.fastq.\\` for \\`.fq.\\` is NOT allowed
+# - omitting \\`.gz\\` is allowed
+
+fastq_regex = r"^([A-Za-z0-9\\\\-_\\\\.]+)_S(\\\\d+)_(L(\\\\d+)_)?[RI](\\\\d+)_(\\\\d+)\\\\.fastq(\\\\.gz)?\\$"
+# assert re.match(fastq_regex, "5k_human_GEX_1_subset_S1_L001_R1_001.fastq.gz") is not None
+# assert re.match(fastq_regex, "5k_human_GEX_1_subset_S1_R1_001.fastq") is not None
+# assert re.match(fastq_regex, "5k_human_GEX_1_subset_S1_R1_001.fastq.gz.txt") is None
+
+# Invert some parameters. Keep the original ones in the config for compatibility
+inverted_params = {
+    "gex_no_secondary_analysis": "gex_secondary_analysis",
+}
+for inverted_param, param in inverted_params.items():
+    par[inverted_param] = not par[param] if par[param] is not None else None
+    del par[param]
+
+GEX_CONFIG_KEYS = {
+    "gex_reference": "reference",
+    "gex_expect_cells": "expect-cells",
+    "gex_force_cells": "force-cells",
+    "gex_chemistry": "chemistry",
+    "gex_no_secondary_analysis": "no-secondary",
+    "gex_generate_bam": "create-bam",
+    "gex_include_introns": "include-introns",
+    "min_assignment_confidence": "min-assignment-confidence",
+    "check_library_compatibility": "check-library-compatibility",
+    "barcode_sample_assignment": "barcode-sample-assignment",
+    "cmo_set": "cmo-set",
+    "probe_set": "probe-set",
+    "filter_probes": "filter-probes",
+    "gex_r1_length": "r1-length",
+    "gex_r2_length": "r2-length",
+    "tenx_cloud_token_path": "tenx-cloud-token-path",
+    "cell_annotation_model": "cell-annotation-model",
+    "emptydrops_minimum_umis": "emptydrops_minimum_umis",
+}
+
+FEATURE_CONFIG_KEYS = {
+    "feature_reference": "reference",
+    "feature_r1_length": "r1-length",
+    "feature_r2_length": "r2-length",
+    "min_crispr_umi": "min-crispr-umi",
+}
+
+VDJ_CONFIG_KEYS = {
+    "vdj_reference": "reference",
+    "vdj_inner_enrichment_primers": "inner-enrichment-primers",
+    "vdj_r1_length": "r1-length",
+    "vdj_r2_length": "r2-length",
+    "vdj_denovo": "denovo",
+}
+
+
+ANTIGEN_SPECIFICITY_CONFIG_KEYS = {
+    "control_id": "control_id",
+    "mhc_allele": "mhc_allele",
+}
+
+
+REFERENCE_SECTIONS = {
+    "gene-expression": (GEX_CONFIG_KEYS, "index"),
+    "feature": (FEATURE_CONFIG_KEYS, "index"),
+    "vdj": (VDJ_CONFIG_KEYS, "index"),
+    "antigen-specificity": (ANTIGEN_SPECIFICITY_CONFIG_KEYS, "columns"),
+}
+
+LIBRARY_CONFIG_KEYS = {
+    "library_id": "fastq_id",
+    "library_type": "feature_types",
+    "library_subsample": "subsample_rate",
+    "library_lanes": "lanes",
+    "library_chemistry": "chemistry",
+}
+
+
+SAMPLE_PARAMS_CONFIG_KEYS = {
+    "sample_ids": "sample_id",
+    "cell_multiplex_oligo_ids": "cmo_ids",
+    "sample_description": "description",
+    "probe_barcode_ids": "probe_barcode_ids",
+    "sample_expect_cells": "expect_cells",
+    "sample_force_cells": "force_cells",
+    "hashtag_ids": "hashtag_ids",
+    "ocm_barcode_ids": "ocm_barcode_ids",
+}
+
+
+# These are derived from the dictionaries above
+REFERENCES = tuple(
+    reference_param
+    for reference_param, cellranger_param in chain(
+        GEX_CONFIG_KEYS.items(), FEATURE_CONFIG_KEYS.items(), VDJ_CONFIG_KEYS.items()
+    )
+    if cellranger_param == "reference"
+)
+LIBRARY_PARAMS = tuple(LIBRARY_CONFIG_KEYS.keys())
+SAMPLE_PARAMS = tuple(SAMPLE_PARAMS_CONFIG_KEYS.keys())
+HELPER_INPUT = {
+    "gex_input": "Gene Expression",
+    "abc_input": "Antibody Capture",
+    "cgc_input": "CRISPR Guide Capture",
+    "mux_input": "Multiplexing Capture",
+    "vdj_input": "VDJ",
+    "vdj_t_input": "VDJ-T",
+    "vdj_t_gd_input": "VDJ-T-GD",
+    "vdj_b_input": "VDJ-B",
+    "agc_input": "Antigen Capture",
+}
+
+
+def infer_library_id_from_path(input_path: str) -> str:
+    match = re.match(fastq_regex, input_path)
+    assert match is not None, (
+        f"File name of '{input_path}' should match regex {fastq_regex}."
+    )
+    return match.group(1)
+
+
+def transform_helper_inputs(par: dict[str, Any]) -> dict[str, Any]:
+    helper_input = {"input": [], "library_id": [], "library_type": []}
+    for input_type, library_type in HELPER_INPUT.items():
+        if par[input_type]:
+            par[input_type] = resolve_input_directories_to_fastq_paths(par[input_type])
+
+            library_ids = [
+                infer_library_id_from_path(path.name) for path in par[input_type]
+            ]
+
+            library_id_dict = {}
+            for fastq, library_id in zip(par[input_type], library_ids):
+                library_id_dict.setdefault(library_id, []).append(fastq)
+
+            for library_id, input in library_id_dict.items():
+                helper_input["input"] += input
+                helper_input["library_id"].append(library_id)
+                helper_input["library_type"].append(library_type)
+
+    assert len(helper_input["library_id"]) == len(set(helper_input["library_id"])), (
+        "File names passed to feature type-specific inputs must be unique"
+    )
+
+    return helper_input
+
+
+def lengths_gt1(dic: dict[str, Optional[list[Any]]]) -> dict[str, int]:
+    return {
+        key: len(li)
+        for key, li in dic.items()
+        if li is not None and isinstance(li, (list, tuple, set))
+    }
+
+
+def strip_margin(text: str) -> str:
+    return re.sub("(\\\\n?)[ \\\\t]*\\\\|", "\\\\\\\\1", text)
+
+
+def subset_dict(
+    dictionary: dict[str, str], keys: Union[dict[str, str], list[str]]
+) -> dict[str, str]:
+    if isinstance(keys, (list, tuple)):
+        keys = {key: key for key in keys}
+    return {
+        dest_key: dictionary[orig_key]
+        for orig_key, dest_key in keys.items()
+        if dictionary[orig_key] is not None
+    }
+
+
+def check_subset_dict_equal_length(
+    group_name: str, dictionary: dict[str, list[str]]
+) -> None:
+    lens = lengths_gt1(dictionary)
+    assert len(set(lens.values())) <= 1, (
+        f"The number of values passed to {group_name} "
+        f"arguments must be 0, 1 or all the same. Offenders: {lens}"
+    )
+
+
+def resolve_input_directories_to_fastq_paths(input_paths: list[str]) -> list[Path]:
+    input_paths = [Path(fastq) for fastq in input_paths]
+    if len(input_paths) == 1 and input_paths[0].is_dir():
+        logger.info(
+            "Detected a directory in input paths, "
+            "traversing to see if we can detect any FASTQ files."
+        )
+        input_paths = [
+            input_path
+            for input_path in input_paths[0].rglob("*")
+            if re.match(fastq_regex, input_path.name)
+        ]
+
+    # check input fastq files
+    for input_path in input_paths:
+        assert re.match(fastq_regex, input_path.name) is not None, (
+            f"File name of --input '{input_path}' should match regex {fastq_regex}."
+        )
+
+    return input_paths
+
+
+def make_paths_absolute(par: dict[str, Any], config: Path | str):
+    with open(config, "r", encoding="utf-8") as open_viash_config:
+        config = yaml.safe_load(open_viash_config)
+
+    arguments = {
+        arg["name"].removeprefix("-").removeprefix("-"): arg
+        for group in config["argument_groups"]
+        for arg in group["arguments"]
+    }
+    for arg_name, arg in arguments.items():
+        if not par.get(arg_name) or arg["type"] != "file":
+            continue
+        par_value, is_multiple = par[arg_name], arg["multiple"]
+        assert is_multiple in (True, False)
+
+        def make_path_absolute(file: str | Path) -> Path:
+            logger.info("Making path %s absolute", file)
+            return Path(file).resolve()
+
+        new_arg = (
+            [make_path_absolute(file) for file in par_value]
+            if is_multiple
+            else make_path_absolute(par_value)
+        )
+        par[arg_name] = new_arg
+    return par
+
+
+def handle_integers_not_set(par: dict[str, Any], viash_config: Path | str) -> str:
+    """
+    Allow to use \\`-1\\` to define a 'not set' value for arguments of \\`type: integer\\` with \\`multiple: true\\`.
+    """
+    with open(viash_config, "r", encoding="utf-8") as open_viash_config:
+        config = yaml.safe_load(open_viash_config)
+
+    arguments = {
+        arg["name"].removeprefix("-").removeprefix("-"): arg
+        for group in config["argument_groups"]
+        for arg in group["arguments"]
+    }
+    for arg_name, arg in arguments.items():
+        if not par.get(arg_name) or arg["type"] != "integer":
+            continue
+        par_value, is_multiple = par[arg_name], arg["multiple"]
+        assert is_multiple in (True, False)
+
+        if not is_multiple:
+            continue
+
+        def replace_notset_values(integer_value: int) -> int | None:
+            return None if integer_value == -1 else integer_value
+
+        # Use an extension array to handle "None" values, otherwise int + NA
+        # values would be converted to a "float" dtype
+        new_arg = pd.array(
+            [replace_notset_values(value) for value in par_value], dtype="Int64"
+        )
+        par[arg_name] = new_arg
+    return par
+
+
+def process_params(par: dict[str, Any], viash_config: Path | str) -> str:
+    if par["input"]:
+        assert len(par["library_type"]) > 0, (
+            "--library_type must be defined when passing input to --input"
+        )
+        assert len(par["library_id"]) > 0, (
+            "--library_id must be defined when passing input to --input"
+        )
+
+        # if par["input"] is a directory, look for fastq files
+        par["input"] = resolve_input_directories_to_fastq_paths(par["input"])
+
+    # add helper input
+    helper_input = transform_helper_inputs(par)
+    for key in ["input", "library_id", "library_type"]:
+        par[key] = (par[key] if par[key] else []) + helper_input[key]
+
+        assert len(par[key]) > 0, (
+            f"Either --{key} or feature type-specific input (e.g. --gex_input, --abc_input, ...) must be defined"
+        )
+
+    # check lengths of libraries metadata
+    library_dict = subset_dict(par, LIBRARY_PARAMS)
+    check_subset_dict_equal_length("Library", library_dict)
+
+    samples_dict = subset_dict(par, SAMPLE_PARAMS)
+    check_subset_dict_equal_length("Samples", samples_dict)
+
+    # Allow using -1 to indicate unset integers for arguments
+    # that accept multiple integers.
+    par = handle_integers_not_set(par, viash_config)
+
+    # use absolute paths
+    return make_paths_absolute(par, viash_config)
+
+
+def generate_csv_category(name: str, args: dict[str, str], orient: str) -> list[str]:
+    assert orient in ("index", "columns")
+    if not args:
+        return []
+    title = [f"[{name}]"]
+    # Which index to include in csv section is based on orientation
+    to_csv_args = {"index": (orient == "index"), "header": (orient == "columns")}
+    values = [pd.DataFrame.from_dict(args, orient=orient).to_csv(**to_csv_args).strip()]
+    return title + values + [""]
+
+
+def generate_config(par: dict[str, Any], fastq_dir: str) -> str:
+    content_list = []
+    par["fastqs"] = fastq_dir
+    libraries = dict(LIBRARY_CONFIG_KEYS, **{"fastqs": "fastqs"})
+    # TODO: use the union (|) operator when python is updated to 3.9
+    all_sections = REFERENCE_SECTIONS | {
+        "libraries": (libraries, "columns"),
+        "samples": (SAMPLE_PARAMS_CONFIG_KEYS, "columns"),
+    }
+    for section_name, (section_params, orientation) in all_sections.items():
+        reference_pars = subset_dict(par, section_params)
+        content_list += generate_csv_category(
+            section_name, reference_pars, orient=orientation
+        )
+
+    return "\\\\n".join(content_list)
+
+
+def untar(tar_file, output_directory):
+    logger.info("Looking at %s to check if it needs decompressing", tar_file)
+    if tar_file and Path(tar_file).is_file() and tarfile.is_tarfile(tar_file):
+        extaction_dir_name = Path(
+            tar_file.stem
+        ).stem  # Remove two extensions (if they exist)
+        unpacked_directory = output_directory / extaction_dir_name
+        logger.info("Extracting %s to %s", tar_file, unpacked_directory)
+
+        with tarfile.open(tar_file, "r") as open_tar:
+            members = open_tar.getmembers()
+            root_dirs = [
+                member
+                for member in members
+                if member.isdir() and member.name != "." and "/" not in member.name
+            ]
+            # if there is only one root_dir (and there are files in that directory)
+            # strip that directory name from the destination folder
+            if len(root_dirs) == 1:
+                for mem in members:
+                    mem.path = Path(*Path(mem.path).parts[1:])
+            members_to_move = [mem for mem in members if mem.path != Path(".")]
+            open_tar.extractall(unpacked_directory, members=members_to_move)
+        return unpacked_directory
+    return tar_file
+
+
+def main(par: dict[str, Any], meta: dict[str, Any]):
+    logger.info("Processing params")
+    par = process_params(par, meta["config"])
+    logger.info(par)
+
+    with tempfile.TemporaryDirectory(
+        prefix="cellranger_multi-", dir=meta["temp_dir"]
+    ) as temp_dir:
+        logger.info("Using temporary directory %s", temp_dir)
+        temp_dir_path = Path(temp_dir)
+        for reference_par_name in REFERENCES:
+            par[reference_par_name] = untar(par[reference_par_name], temp_dir_path)
+
+        logger.info("Creating symbolic links to temporary directory")
+        # Creating symlinks of fastq files to tempdir
+        input_symlinks_dir = temp_dir_path / "input_symlinks"
+        input_symlinks_dir.mkdir()
+        for fastq in par["input"]:
+            destination = input_symlinks_dir / fastq.name
+            destination.symlink_to(fastq)
+
+        logger.info("Creating config file")
+        config_content = generate_config(par, input_symlinks_dir)
+        logger.info("Using config: \\\\n\\\\t%s", config_content.replace("\\\\n", "\\\\n\\\\t"))
+        par["output"].mkdir(parents=True, exist_ok=True)
+        config_file = par["output"] / "config.csv"
+        with open(config_file, "w") as f:
+            f.write(config_content)
+
+        logger.info("Creating Cell Ranger argument list")
+        temp_id = "run"
+        cmd = [CELL_RANGER_EXEC, "multi", "--disable-ui", "--id", temp_id]
+
+        command_line_parameters = {
+            "--localcores": int(meta["cpus"]) if meta["cpus"] else cpu_count() - 1,
+            "--csv": config_file,
+        }
+        if meta["memory_gb"]:
+            command_line_parameters["--localmem"] = (
+                int(meta["memory_gb"]) - 2
+                if meta["memory_gb"] > 2
+                else meta["memory_gb"]
+            )
+        cmd += [
+            f"{param}={param_value}"
+            for param, param_value in command_line_parameters.items()
+        ]
+
+        logger.info("> " + " ".join(cmd))
+
+        if par["dryrun"]:
+            logger.info("Not executing command. Dry run enabled.")
+            return
+
+        logger.info("Starting Cell Ranger")
+        with (
+            subprocess.Popen(
+                cmd,
+                text=True,
+                stdout=subprocess.PIPE,
+                stderr=subprocess.STDOUT,
+                bufsize=1,
+                cwd=temp_dir,
+            ) as process,
+            (par["output"] / "cellranger_multi.log").open("w") as open_log,
+        ):
+            for line in process.stdout:
+                print(line.strip(), flush=True)
+                open_log.write(line)
+        logger.info("Checking exit code.")
+        # Popen objects do not have a check_exitcode method.
+        if process.returncode != 0:
+            raise RuntimeError("Cell Ranger returned a nonzero exitcode!")
+
+        logger.info("Checking if expected output files are present.")
+        # Sanity check the output: check for required files
+        tmp_output_dir = temp_dir_path / temp_id / "outs"
+        expected_files = {
+            Path("per_sample_outs"): Path.is_dir,
+            Path("config.csv"): Path.is_file,
+            Path("filtered_feature_bc_matrix"): Path.is_dir,
+            Path("filtered_feature_bc_matrix.h5"): Path.is_file,
+            Path("raw_feature_bc_matrix"): Path.is_dir,
+            Path("raw_feature_bc_matrix.h5"): Path.is_file,
+        }
+        dir_contents = map(str, tmp_output_dir.iterdir())
+        logger.info("Directory contents: %s", ", ".join(dir_contents))
+        for file_path, type_func in expected_files.items():
+            output_path = tmp_output_dir / file_path
+            if not type_func(output_path):
+                raise ValueError(f"Could not find expected '{output_path}'")
+        logger.info("Copying output files to %s", par["output"])
+        for output_path in tmp_output_dir.rglob("*"):
+            if output_path.name != "config.csv":  # Already created
+                shutil.move(str(output_path), par["output"])
+
+
+if __name__ == "__main__":
+    logger.info("Component '%s' started.", meta["name"])
+    main(par, meta)
+    logger.info("Finished!")
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/mapping/cellranger_multi",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "veryhighmem",
+    "highcpu",
+    "highdisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/nextflow.config
new file mode 100644
index 0000000..8902a5a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'mapping/cellranger_multi'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Align fastq files using Cell Ranger multi.'
+  author = 'Dries Schaumont, Angela Oliveira Pisco, Robrecht Cannoodt, Dries De Maeyer, Weiwei Schultz, Dorien Roosen'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/nextflow_schema.json
new file mode 100644
index 0000000..2b41b9a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/nextflow_schema.json
@@ -0,0 +1,579 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "cellranger_multi",
+  "description": "Align fastq files using Cell Ranger multi.",
+  "type": "object",
+  "$defs": {
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "The folder to store the alignment results.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output\"`, direction: `output`, example: `\"/path/to/output\"`. ",
+          "default": "$id.$key.output"
+        }
+      }
+    },
+    "input files": {
+      "title": "Input files",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        }
+      }
+    },
+    "feature type-specific input files": {
+      "title": "Feature type-specific input files",
+      "type": "object",
+      "description": "Helper functionality to allow feature type-specific input files, without the need to specify\nlibrary_type or library_id. The library_id will be inferred from the input paths.\n",
+      "properties": {
+        "gex_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for Gene Expression",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "abc_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for Antibody Capture",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "cgc_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for CRISPR Guide Capture",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "mux_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for Multiplexing Capture",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "vdj_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for VDJ",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "vdj_t_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for VDJ-T",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "vdj_t_gd_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for VDJ-T-GD",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "vdj_b_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for VDJ-B",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "agc_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for Antigen Capture",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        }
+      }
+    },
+    "library arguments": {
+      "title": "Library arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "library_id": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The Illumina sample name to analyze",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"mysample1\"]`. "
+        },
+        "library_type": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The underlying feature type of the library.\n",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"Gene Expression\"]`, choices: ``Gene Expression`, `VDJ`, `VDJ-T`, `VDJ-B`, `VDJ-T-GD`, `Antibody Capture`, `CRISPR Guide Capture`, `Multiplexing Capture`, `Antigen Capture`, `Custom``. "
+        },
+        "library_subsample": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The rate at which reads from the provided FASTQ files are sampled.\nMust be strictly greater than 0 and less than or equal to 1.\n",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"0.5\"]`. "
+        },
+        "library_lanes": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Lanes associated with this sample",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"1-4\"]`. "
+        },
+        "library_chemistry": {
+          "type": "string",
+          "description": "Only applicable to FRP",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "sample parameters": {
+      "title": "Sample parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "sample_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "A name to identify a multiplexed sample",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "sample_description": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "A description for the sample.",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "sample_expect_cells": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Expected number of recovered cells, used as input to cell calling algorithm.\n",
+          "help_text": "Type: `integer`, multiple: `True`, example: `[3000]`. "
+        },
+        "sample_force_cells": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Force pipeline to use this number of cells, bypassing cell detection.\n",
+          "help_text": "Type: `integer`, multiple: `True`, example: `[3000]`. "
+        }
+      }
+    },
+    "feature barcode library specific arguments": {
+      "title": "Feature Barcode library specific arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "feature_reference": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to the Feature reference CSV file, declaring Feature Barcode constructs and associated barcodes.\nRequired only for Antibody Capture or CRISPR Guide Capture libraries.\nSee https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/using/feature-bc-analysis#feature-ref for more information.\"\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"feature_reference.csv\"`. "
+        },
+        "feature_r1_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,\nwhere N is the user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "feature_r2_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,\nwhere N is a user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "min_crispr_umi": {
+          "type": "integer",
+          "description": "Set the minimum number of CRISPR guide RNA UMIs required for protospacer detection.\nIf a lower or higher sensitivity is desired for detection, this value can be customized\naccording to specific experimental needs",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        }
+      }
+    },
+    "gene expression arguments": {
+      "title": "Gene expression arguments",
+      "type": "object",
+      "description": "Arguments relevant to the analysis of gene expression data.",
+      "properties": {
+        "gex_reference": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Genome refence index built by Cell Ranger mkref.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"reference_genome.tar.gz\"`. "
+        },
+        "gex_secondary_analysis": {
+          "type": "boolean",
+          "description": "Whether or not to run the secondary analysis e.g",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "gex_generate_bam": {
+          "type": "boolean",
+          "description": "Whether to generate a BAM file.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "tenx_cloud_token_path": {
+          "type": "string",
+          "format": "path",
+          "description": "The 10x Cloud Analysis user token used to enable cell annotation.",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`. "
+        },
+        "cell_annotation_model": {
+          "type": "string",
+          "description": "\"Cell annotation model to use",
+          "help_text": "Type: `string`, multiple: `False`, choices: ``auto`, `human_pca_v1_beta`, `mouse_pca_v1_beta``. ",
+          "enum": [
+            "auto",
+            "human_pca_v1_beta",
+            "mouse_pca_v1_beta"
+          ]
+        },
+        "gex_expect_cells": {
+          "type": "integer",
+          "description": "Expected number of recovered cells, used as input to cell calling algorithm.\n",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3000`. "
+        },
+        "gex_force_cells": {
+          "type": "integer",
+          "description": "Force pipeline to use this number of cells, bypassing cell detection.\n",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3000`. "
+        },
+        "gex_include_introns": {
+          "type": "boolean",
+          "description": "Whether or not to include intronic reads in counts.\nThis option does not apply to Fixed RNA Profiling analysis.\n",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "gex_r1_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,\nwhere N is the user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "gex_r2_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,\nwhere N is a user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "gex_chemistry": {
+          "type": "string",
+          "description": "Assay configuration",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"auto\"`, choices: ``auto`, `threeprime`, `fiveprime`, `SC3Pv1`, `SC3Pv2`, `SC3Pv3`, `SC3Pv3-polyA`, `SC3Pv4`, `SC3Pv4-polyA`, `SC3Pv3LT`, `SC3Pv3HT`, `SC3Pv3HT-polyA`, `SC5P-PE`, `SC5P-PE-v3`, `SC5P-R2`, `SC-FB`, `SC5P-R2-v3`, `SCP5-PE-v3`, `SC5PHT`, `MFRP`, `MFRP-R1`, `MFRP-RNA`, `MFRP-Ab`, `SFRP`, `MFRP-Ab-R2pos50`, `MFRP-RNA-R1`, `MFRP-Ab-R1`, `ARC-v1``. ",
+          "enum": [
+            "auto",
+            "threeprime",
+            "fiveprime",
+            "SC3Pv1",
+            "SC3Pv2",
+            "SC3Pv3",
+            "SC3Pv3-polyA",
+            "SC3Pv4",
+            "SC3Pv4-polyA",
+            "SC3Pv3LT",
+            "SC3Pv3HT",
+            "SC3Pv3HT-polyA",
+            "SC5P-PE",
+            "SC5P-PE-v3",
+            "SC5P-R2",
+            "SC-FB",
+            "SC5P-R2-v3",
+            "SCP5-PE-v3",
+            "SC5PHT",
+            "MFRP",
+            "MFRP-R1",
+            "MFRP-RNA",
+            "MFRP-Ab",
+            "SFRP",
+            "MFRP-Ab-R2pos50",
+            "MFRP-RNA-R1",
+            "MFRP-Ab-R1",
+            "ARC-v1"
+          ],
+          "default": "auto"
+        }
+      }
+    },
+    "vdj related parameters": {
+      "title": "VDJ related parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "vdj_reference": {
+          "type": "string",
+          "format": "path",
+          "description": "VDJ refence index built by Cell Ranger mkref.",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"reference_vdj.tar.gz\"`. "
+        },
+        "vdj_inner_enrichment_primers": {
+          "type": "string",
+          "format": "path",
+          "description": "V(D)J Immune Profiling libraries: if inner enrichment primers other than those provided \nin the 10x Genomics kits are used, they need to be specified here as a\ntext file with one primer per line.\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"enrichment_primers.txt\"`. "
+        },
+        "vdj_r1_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases, where N is the user-supplied value.\nNote that the length includes the Barcode and UMI sequences so do not set this below 26.\n",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "vdj_r2_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases, where N is a user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "vdj_denovo": {
+          "type": "boolean",
+          "description": "Run in reference-free mode (i.e., do not use annotations)",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        }
+      }
+    },
+    "3' cell multiplexing parameters (cellplex multiplexing)": {
+      "title": "3' Cell multiplexing parameters (CellPlex Multiplexing)",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "cell_multiplex_oligo_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The Cell Multiplexing oligo IDs used to multiplex this sample",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "min_assignment_confidence": {
+          "type": "number",
+          "description": "The minimum estimated likelihood to call a sample as tagged with a Cell Multiplexing Oligo (CMO) instead of \"Unassigned\".\nUsers may wish to tolerate a higher rate of mis-assignment in order to obtain more singlets to include in their analysis,\nor a lower rate of mis-assignment at the cost of obtaining fewer singlets.\n",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "cmo_set": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to a custom CMO set CSV file, declaring CMO constructs and associated barcodes",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`. "
+        },
+        "barcode_sample_assignment": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to a barcode-sample assignment CSV file that specifies the barcodes that belong to each sample.\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`. "
+        }
+      }
+    },
+    "hashtag multiplexing parameters": {
+      "title": "Hashtag multiplexing parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "hashtag_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The hashtag IDs used to multiplex this sample",
+          "help_text": "Type: `string`, multiple: `True`. "
+        }
+      }
+    },
+    "on-chip multiplexing parameters": {
+      "title": "On-chip multiplexing parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "ocm_barcode_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The OCM barcode IDs used to multiplex this sample",
+          "help_text": "Type: `string`, multiple: `True`. "
+        }
+      }
+    },
+    "flex multiplexing paramaters": {
+      "title": "Flex multiplexing paramaters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "probe_set": {
+          "type": "string",
+          "format": "path",
+          "description": "A probe set reference CSV file",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`. "
+        },
+        "filter_probes": {
+          "type": "boolean",
+          "description": "If 'false', include all non-deprecated probes listed in the probe set reference CSV file.\nIf 'true' or not set, probes that are predicted to have off-target activity to homologous genes are excluded from analysis.\nNot filtering will result in UMI counts from all non-deprecated probes,\nincluding those with predicted off-target activity, to be used in the analysis.\nProbes whose ID is prefixed with DEPRECATED are always excluded from the analysis.\n",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        },
+        "probe_barcode_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The Fixed RNA Probe Barcode ID used for this sample, and for multiplex GEX + Antibody Capture libraries,\nthe corresponding Antibody Multiplexing Barcode IDs",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "emptydrops_minimum_umis": {
+          "type": "integer",
+          "description": "For singleplex Flex experiments, use this option to adjust the UMI cutoff during the second step of cell calling.\nCell Ranger will still perform the full cell calling process but will only evaluate barcodes with UMIs above\nthe threshold you specify.\n",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        }
+      }
+    },
+    "antigen capture (beam) libary arguments": {
+      "title": "Antigen Capture (BEAM) libary arguments",
+      "type": "object",
+      "description": "These arguments are recommended if an Antigen Capture (BEAM) library is present. \nIt is needed to calculate the antigen specificity score.\n",
+      "properties": {
+        "control_id": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "A user-defined ID for any negative controls used in the T/BCR Antigen Capture assay",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "mhc_allele": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The MHC allele for TCR Antigen Capture libraries",
+          "help_text": "Type: `string`, multiple: `True`. "
+        }
+      }
+    },
+    "general arguments": {
+      "title": "General arguments",
+      "type": "object",
+      "description": "These arguments are applicable to all library types.\n",
+      "properties": {
+        "check_library_compatibility": {
+          "type": "boolean",
+          "description": "Optional",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "executor arguments": {
+      "title": "Executor arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "dryrun": {
+          "type": "boolean",
+          "description": "If true, the output directory will only contain the CWL input files, but the pipeline itself will not be executed.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/input files"
+    },
+    {
+      "$ref": "#/$defs/feature type-specific input files"
+    },
+    {
+      "$ref": "#/$defs/library arguments"
+    },
+    {
+      "$ref": "#/$defs/sample parameters"
+    },
+    {
+      "$ref": "#/$defs/feature barcode library specific arguments"
+    },
+    {
+      "$ref": "#/$defs/gene expression arguments"
+    },
+    {
+      "$ref": "#/$defs/vdj related parameters"
+    },
+    {
+      "$ref": "#/$defs/3' cell multiplexing parameters (cellplex multiplexing)"
+    },
+    {
+      "$ref": "#/$defs/hashtag multiplexing parameters"
+    },
+    {
+      "$ref": "#/$defs/on-chip multiplexing parameters"
+    },
+    {
+      "$ref": "#/$defs/flex multiplexing paramaters"
+    },
+    {
+      "$ref": "#/$defs/antigen capture (beam) libary arguments"
+    },
+    {
+      "$ref": "#/$defs/general arguments"
+    },
+    {
+      "$ref": "#/$defs/executor arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/mapping/cellranger_multi/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/.config.vsh.yaml
new file mode 100644
index 0000000..b94acc7
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/.config.vsh.yaml
@@ -0,0 +1,274 @@
+name: "add_id"
+namespace: "metadata"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to the input .h5mu."
+    info: null
+    example:
+    - "sample_path"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_id"
+    description: "The input id."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_output"
+    description: "Name of the .obs column where to store the id."
+    info: null
+    default:
+    - "sample_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Name of output MuData file.\n"
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--make_observation_keys_unique"
+    description: "Join the id to the .obs index (.obs_names)."
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Add id of .obs. Also allows to make .obs_names (the .obs index) unique\
+  \ \nby prefixing the values with an unique id per .h5mu file.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "lowmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.11-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/metadata/add_id/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/metadata/add_id"
+  executable: "target/nextflow/metadata/add_id/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/main.nf
new file mode 100644
index 0000000..7855d30
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/main.nf
@@ -0,0 +1,3997 @@
+// add_id v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "add_id",
+  "namespace" : "metadata",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to the input .h5mu.",
+          "example" : [
+            "sample_path"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_id",
+          "description" : "The input id.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_output",
+          "description" : "Name of the .obs column where to store the id.",
+          "default" : [
+            "sample_id"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Name of output MuData file.\n",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--make_observation_keys_unique",
+          "description" : "Join the id to the .obs index (.obs_names).",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Add id of .obs. Also allows to make .obs_names (the .obs index) unique \nby prefixing the values with an unique id per .h5mu file.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data/e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "lowmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.11-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/metadata/add_id/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/metadata/add_id",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+from __future__ import annotations
+import sys
+from mudata import read_h5mu, MuData
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_id': $( if [ ! -z ${VIASH_PAR_INPUT_ID+x} ]; then echo "r'${VIASH_PAR_INPUT_ID//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_output': $( if [ ! -z ${VIASH_PAR_OBS_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OBS_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'make_observation_keys_unique': $( if [ ! -z ${VIASH_PAR_MAKE_OBSERVATION_KEYS_UNIQUE+x} ]; then echo "r'${VIASH_PAR_MAKE_OBSERVATION_KEYS_UNIQUE//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+
+def make_observation_keys_unique(sample_id: str, sample: MuData) -> None:
+    """
+    Make the observation keys unique across all samples. At input,
+    the observation keys are unique within a sample. By adding the sample name
+    (unique for a sample) to each observation key, the observation key is made
+    unique across all samples as well.
+    """
+    logger.info(
+        "Making observation keys unique across all "
+        "samples by appending prefix '%s' to the observation names.",
+        sample_id,
+    )
+    sample.obs.index = f"{sample_id}_" + sample.obs.index
+    make_observation_keys_unique_per_mod(sample_id, sample)
+    logger.info("Done making observation keys unique.")
+
+
+def make_observation_keys_unique_per_mod(sample_id: str, sample: MuData) -> None:
+    """
+    Updating MuData.obs_names is not allowed (it is read-only).
+    So the observation keys for each modality has to be updated manually.
+    """
+    for mod_name, mod in sample.mod.items():
+        logger.info("Processing modality '%s'", mod_name)
+        mod.obs_names = f"{sample_id}_" + mod.obs_names
+
+
+def main():
+    logger.info("Reading input file '%s'.", par["input"])
+    input_data = read_h5mu(par["input"])
+    logger.info(
+        "Adding column '%s' to global .obs dataframe, populated with ID '%s'",
+        par["obs_output"],
+        par["input_id"],
+    )
+    input_data.obs[par["obs_output"]] = par["input_id"]
+    logger.info("Done adding column to global .obs")
+    for mod_name, mod_data in input_data.mod.items():
+        logger.info(
+            "Adding column '%s' to .obs dataframe for modality '%s', "
+            "populated with ID '%s'",
+            par["obs_output"],
+            mod_name,
+            par["input_id"],
+        )
+        mod_data.obs[par["obs_output"]] = par["input_id"]
+    logger.info("Done adding per-modality columns.")
+    if par["make_observation_keys_unique"]:
+        make_observation_keys_unique(par["input_id"], input_data)
+    logger.info(
+        "Writing out data to '%s' with compression '%s'.",
+        par["output"],
+        par["output_compression"],
+    )
+    input_data.write_h5mu(par["output"], compression=par["output_compression"])
+    logger.info("Finished")
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/metadata/add_id",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "singlecpu",
+    "lowmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/nextflow.config
new file mode 100644
index 0000000..9282f17
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'metadata/add_id'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Add id of .obs. Also allows to make .obs_names (the .obs index) unique \nby prefixing the values with an unique id per .h5mu file.\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/nextflow_schema.json
new file mode 100644
index 0000000..83fcf94
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/nextflow_schema.json
@@ -0,0 +1,75 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "add_id",
+  "description": "Add id of .obs. Also allows to make .obs_names (the .obs index) unique \nby prefixing the values with an unique id per .h5mu file.\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the input .h5mu.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"sample_path\"`. "
+        },
+        "input_id": {
+          "type": "string",
+          "description": "The input id.",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        },
+        "obs_output": {
+          "type": "string",
+          "description": "Name of the .obs column where to store the id.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"sample_id\"`. ",
+          "default": "sample_id"
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Name of output MuData file.\n",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "make_observation_keys_unique": {
+          "type": "boolean",
+          "description": "Join the id to the .obs index (.obs_names).",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/add_id/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/.config.vsh.yaml
new file mode 100644
index 0000000..0974e6a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/.config.vsh.yaml
@@ -0,0 +1,330 @@
+name: "grep_annotation_column"
+namespace: "metadata"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  description: "Arguments related to the input dataset."
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to the input .h5mu."
+    info: null
+    example:
+    - "sample_path"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_column"
+    description: "Column to query. If not specified, use .var_names or .obs_names,\
+      \ depending on the value of --matrix"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_layer"
+    description: "Input data to use when calculating fraction of observations that\
+      \ match with the query. \nOnly used when --output_fraction_column is provided.\
+      \ If not specified, .X is used.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality to get the annotation matrix from.\n"
+    info: null
+    example:
+    - "rna"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--matrix"
+    description: "Matrix to fetch the column from that will be searched."
+    info: null
+    example:
+    - "var"
+    required: false
+    choices:
+    - "var"
+    - "obs"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  description: "Arguments related to how the output will be written."
+  arguments:
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Location of the output MuData file.\n"
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_match_column"
+    description: "Name of the column to write the result to."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_fraction_column"
+    description: "For the opposite axis, name of the column to write the fraction\
+      \ of \nobservations that matches to the pattern.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Query options"
+  description: "Options related to the query"
+  arguments:
+  - type: "string"
+    name: "--regex_pattern"
+    description: "Regex to use to match with the input column."
+    info: null
+    example:
+    - "^[mM][tT]-"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Perform a regex lookup on a column from the annotation matrices .obs\
+  \ or .var.\nThe annotation matrix can originate from either a modality, or all modalities\
+  \ (global .var or .obs).\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "lowmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.11-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/metadata/grep_annotation_column/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/metadata/grep_annotation_column"
+  executable: "target/nextflow/metadata/grep_annotation_column/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/main.nf
new file mode 100644
index 0000000..ad1021a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/main.nf
@@ -0,0 +1,4136 @@
+// grep_annotation_column v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "grep_annotation_column",
+  "namespace" : "metadata",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "description" : "Arguments related to the input dataset.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to the input .h5mu.",
+          "example" : [
+            "sample_path"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_column",
+          "description" : "Column to query. If not specified, use .var_names or .obs_names, depending on the value of --matrix",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_layer",
+          "description" : "Input data to use when calculating fraction of observations that match with the query. \nOnly used when --output_fraction_column is provided. If not specified, .X is used.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality to get the annotation matrix from.\n",
+          "example" : [
+            "rna"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--matrix",
+          "description" : "Matrix to fetch the column from that will be searched.",
+          "example" : [
+            "var"
+          ],
+          "required" : false,
+          "choices" : [
+            "var",
+            "obs"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "description" : "Arguments related to how the output will be written.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Location of the output MuData file.\n",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_match_column",
+          "description" : "Name of the column to write the result to.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_fraction_column",
+          "description" : "For the opposite axis, name of the column to write the fraction of \nobservations that matches to the pattern.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Query options",
+      "description" : "Options related to the query",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--regex_pattern",
+          "description" : "Regex to use to match with the input column.",
+          "example" : [
+            "^[mM][tT]-"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Perform a regex lookup on a column from the annotation matrices .obs or .var.\nThe annotation matrix can originate from either a modality, or all modalities (global .var or .obs).\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data/e18_mouse_brain_fresh_5k_filtered_feature_bc_matrix_subset_unique_obs.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "lowmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.11-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/metadata/grep_annotation_column/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/metadata/grep_annotation_column",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+import mudata as mu
+from pathlib import Path
+from operator import attrgetter
+from pandas import Series
+import scipy as sc
+import re
+import numpy as np
+
+
+### VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_column': $( if [ ! -z ${VIASH_PAR_INPUT_COLUMN+x} ]; then echo "r'${VIASH_PAR_INPUT_COLUMN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_layer': $( if [ ! -z ${VIASH_PAR_INPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_INPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'matrix': $( if [ ! -z ${VIASH_PAR_MATRIX+x} ]; then echo "r'${VIASH_PAR_MATRIX//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_match_column': $( if [ ! -z ${VIASH_PAR_OUTPUT_MATCH_COLUMN+x} ]; then echo "r'${VIASH_PAR_OUTPUT_MATCH_COLUMN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_fraction_column': $( if [ ! -z ${VIASH_PAR_OUTPUT_FRACTION_COLUMN+x} ]; then echo "r'${VIASH_PAR_OUTPUT_FRACTION_COLUMN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'regex_pattern': $( if [ ! -z ${VIASH_PAR_REGEX_PATTERN+x} ]; then echo "r'${VIASH_PAR_REGEX_PATTERN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+### VIASH END
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+
+def describe_array(arr, msg):
+    # Note: sc.stats returns a DescribeResult NamedTuple. For NamedTuples,
+    # the _asdict method is public facing even though it starts with an underscore.
+    description = sc.stats.describe(arr)._asdict()
+    logger.info(
+        "%s:\\\\nshape: %s\\\\nmean: %s\\\\nnobs: %s\\\\n"
+        "variance: %s\\\\nmin: %s\\\\nmax: %s\\\\ncontains na: %s\\\\ndtype: %s\\\\ncontains 0: %s",
+        msg,
+        arr.shape,
+        description["mean"],
+        description["nobs"],
+        description["variance"],
+        description["minmax"][0],
+        description["minmax"][1],
+        np.isnan(arr).any(),
+        arr.dtype,
+        (arr == 0).any(),
+    )
+
+
+def main(par):
+    input_file, output_file, mod_name = (
+        Path(par["input"]),
+        Path(par["output"]),
+        par["modality"],
+    )
+    logger.info(f"Compiling regular expression '{par['regex_pattern']}'.")
+    try:
+        compiled_regex = re.compile(par["regex_pattern"])
+    except (TypeError, re.error) as e:
+        raise ValueError(
+            f"{par['regex_pattern']} is not a valid regular expression pattern."
+        ) from e
+    else:
+        if compiled_regex.groups:
+            raise NotImplementedError(
+                "Using match groups is not supported by this component."
+            )
+    logger.info("Reading input file %s, modality %s.", input_file, mod_name)
+
+    modality_data = mu.read_h5ad(input_file, mod=mod_name)
+    logger.info("Reading input file done.")
+    logger.info("Using annotation dataframe '%s'.", par["matrix"])
+    annotation_matrix = getattr(modality_data, par["matrix"])
+    default_column = {"var": attrgetter("var_names"), "obs": attrgetter("obs_names")}
+    if par["input_column"]:
+        logger.info("Input column '%s' was specified.", par["input_column"])
+        try:
+            annotation_column = annotation_matrix[par["input_column"]]
+        except KeyError as e:
+            raise ValueError(
+                f"Column {par['input_column']} could not be found for modality "
+                f"{par['modality']}. Available columns:"
+                f" {','.join(annotation_matrix.columns.to_list())}"
+            ) from e
+    else:
+        logger.info(f"No input column specified, using '.{par['matrix']}_names'")
+        annotation_column = default_column[par["matrix"]](modality_data).to_series()
+    logger.info("Applying regex search.")
+    grep_result = annotation_column.str.contains(par["regex_pattern"], regex=True)
+    logger.info("Search results: %s", grep_result.value_counts())
+    # A Series object cannot be used as an indexer for a scipy sparse array
+    # when the data type is a pandas boolean extension array because
+    # extension arrays do not define .nonzero()
+    # See https://github.com/pandas-dev/pandas/issues/46025
+    grep_result = grep_result.to_numpy(dtype="bool", na_value=False)
+
+    other_axis_attribute = {"var": "obs", "obs": "var"}
+    if par["output_fraction_column"]:
+        logger.info(
+            "Enabled writing the fraction of values that matches to the pattern."
+        )
+        input_layer = (
+            modality_data.X
+            if not par["input_layer"]
+            else modality_data.layers[par["input_layer"]]
+        )
+        totals = np.ravel(input_layer.sum(axis=1))
+        describe_array(totals, "Summary of total counts for layer")
+        counts_for_matches = np.ravel(input_layer[:, grep_result].sum(axis=1))
+        describe_array(counts_for_matches, "Summary of counts matching grep")
+        with np.errstate(all="raise"):
+            pct_matching = np.divide(
+                counts_for_matches,
+                totals,
+                out=np.zeros_like(totals, dtype=np.float64),
+                where=(~np.isclose(totals, np.zeros_like(totals))),
+            )
+        logger.info("Testing wether or not fractions data contains NA.")
+        assert ~np.isnan(pct_matching).any(), "Fractions should not contain NA."
+        logger.info("Fraction statistics: \\\\n%s", Series(pct_matching).describe())
+        pct_matching = np.where(np.isclose(pct_matching, 0, atol=1e-6), 0, pct_matching)
+        pct_matching = np.where(np.isclose(pct_matching, 1, atol=1e-6), 1, pct_matching)
+        assert (np.logical_and(pct_matching >= 0, pct_matching <= 1)).all(), (
+            "Fractions are not within bounds, please report this as a bug"
+        )
+        output_matrix = other_axis_attribute[par["matrix"]]
+        logger.info(
+            "Writing fractions to matrix '%s', column '%s'",
+            output_matrix,
+            par["output_fraction_column"],
+        )
+        getattr(modality_data, output_matrix)[par["output_fraction_column"]] = (
+            pct_matching
+        )
+    logger.info(
+        "Adding values that matched the pattern to '%s', column '%s'",
+        par["matrix"],
+        par["output_match_column"],
+    )
+    getattr(modality_data, par["matrix"])[par["output_match_column"]] = grep_result
+    logger.info(
+        "Writing out data to '%s' with compression '%s'.",
+        output_file,
+        par["output_compression"],
+    )
+    write_h5ad_to_h5mu_with_compression(
+        output_file, par["input"], mod_name, modality_data, par["output_compression"]
+    )
+
+
+if __name__ == "__main__":
+    main(par)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/metadata/grep_annotation_column",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "singlecpu",
+    "lowmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/nextflow.config
new file mode 100644
index 0000000..c21f9b7
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'metadata/grep_annotation_column'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Perform a regex lookup on a column from the annotation matrices .obs or .var.\nThe annotation matrix can originate from either a modality, or all modalities (global .var or .obs).\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/nextflow_schema.json
new file mode 100644
index 0000000..ff552cb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/nextflow_schema.json
@@ -0,0 +1,117 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "grep_annotation_column",
+  "description": "Perform a regex lookup on a column from the annotation matrices .obs or .var.\nThe annotation matrix can originate from either a modality, or all modalities (global .var or .obs).\n",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "Arguments related to the input dataset.",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the input .h5mu.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"sample_path\"`. "
+        },
+        "input_column": {
+          "type": "string",
+          "description": "Column to query",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "input_layer": {
+          "type": "string",
+          "description": "Input data to use when calculating fraction of observations that match with the query",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality to get the annotation matrix from.\n",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"rna\"`. "
+        },
+        "matrix": {
+          "type": "string",
+          "description": "Matrix to fetch the column from that will be searched.",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"var\"`, choices: ``var`, `obs``. ",
+          "enum": [
+            "var",
+            "obs"
+          ]
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "Arguments related to how the output will be written.",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Location of the output MuData file.\n",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "output_match_column": {
+          "type": "string",
+          "description": "Name of the column to write the result to.",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        },
+        "output_fraction_column": {
+          "type": "string",
+          "description": "For the opposite axis, name of the column to write the fraction of \nobservations that matches to the pattern.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "query options": {
+      "title": "Query options",
+      "type": "object",
+      "description": "Options related to the query",
+      "properties": {
+        "regex_pattern": {
+          "type": "string",
+          "description": "Regex to use to match with the input column.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"^[mM][tT]-\"`. "
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/query options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/grep_annotation_column/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/.config.vsh.yaml
new file mode 100644
index 0000000..7787e6e
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/.config.vsh.yaml
@@ -0,0 +1,272 @@
+name: "move_obsm_to_obs"
+namespace: "metadata"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "MuData Input"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_key"
+    description: "Key of a data structure to move from `.obsm` to `.obs`."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "MuData Output"
+  arguments:
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Move a matrix from .obsm to .obs. Newly created columns in .obs will\
+  \ \nbe created from the .obsm key suffixed with an underscore and the name of the\
+  \ columns\nof the specified .obsm matrix.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "lowmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/metadata/move_obsm_to_obs/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/metadata/move_obsm_to_obs"
+  executable: "target/nextflow/metadata/move_obsm_to_obs/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/main.nf
new file mode 100644
index 0000000..ece06be
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/main.nf
@@ -0,0 +1,3986 @@
+// move_obsm_to_obs v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "move_obsm_to_obs",
+  "namespace" : "metadata",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "MuData Input",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_key",
+          "description" : "Key of a data structure to move from `.obsm` to `.obs`.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "MuData Output",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Move a matrix from .obsm to .obs. Newly created columns in .obs will \nbe created from the .obsm key suffixed with an underscore and the name of the columns\nof the specified .obsm matrix.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "lowmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/metadata/move_obsm_to_obs/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/metadata/move_obsm_to_obs",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+from functools import partial
+from pandas.errors import MergeError
+from mudata import read_h5ad
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_key': $( if [ ! -z ${VIASH_PAR_OBSM_KEY+x} ]; then echo "r'${VIASH_PAR_OBSM_KEY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+logger.info("Reading modality %s from file %s", par["modality"], par["input"])
+try:
+    mod_data = read_h5ad(par["input"], mod=par["modality"])
+except KeyError:
+    raise ValueError(f"Modality {par['modality']} does not exist.")
+
+logger.info("Moving .obm key %s", par["obsm_key"])
+try:
+    obsm_matrix = mod_data.obsm[par["obsm_key"]].copy()
+except KeyError:
+    raise ValueError(
+        f".obsm key {par['obsm_key']} was not found in "
+        f".obsm slot for modality {par['modality']}."
+    )
+
+
+obsm_matrix.rename(
+    partial("{key}_{}".format, key=par["obsm_key"]),
+    axis="columns",
+    copy=False,
+    inplace=True,
+)
+
+original_n_obs = len(mod_data.obs)
+try:
+    logger.info(f".obs names: {mod_data.obs_names}")
+    logger.info(f".obsm index: {obsm_matrix.index}")
+    new_obs = mod_data.obs.drop(obsm_matrix.columns, axis=1, errors="ignore")
+    new_obs = new_obs.merge(
+        obsm_matrix,
+        how="left",
+        validate="one_to_one",
+        left_index=True,
+        right_index=True,
+    )
+    mod_data.obs = new_obs
+except MergeError:
+    raise ValueError(
+        f"Could not join .obsm matrix at {par['obsm_key']} to .obs because there "
+        "are some observation that are not overlapping between the two matrices "
+        "(indexes should overlap). This is either a bug or your mudata file is corrupt."
+    )
+del mod_data.obsm[par["obsm_key"]]
+
+logger.info(
+    "Write output to %s with compression %s", par["output"], par["output_compression"]
+)
+write_h5ad_to_h5mu_with_compression(
+    par["output"], par["input"], par["modality"], mod_data, par["output_compression"]
+)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/metadata/move_obsm_to_obs",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "singlecpu",
+    "lowmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/nextflow.config
new file mode 100644
index 0000000..22fb4f2
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'metadata/move_obsm_to_obs'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Move a matrix from .obsm to .obs. Newly created columns in .obs will \nbe created from the .obsm key suffixed with an underscore and the name of the columns\nof the specified .obsm matrix.\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/nextflow_schema.json
new file mode 100644
index 0000000..44f1383
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/nextflow_schema.json
@@ -0,0 +1,79 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "move_obsm_to_obs",
+  "description": "Move a matrix from .obsm to .obs. Newly created columns in .obs will \nbe created from the .obsm key suffixed with an underscore and the name of the columns\nof the specified .obsm matrix.\n",
+  "type": "object",
+  "$defs": {
+    "mudata input": {
+      "title": "MuData Input",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "obsm_key": {
+          "type": "string",
+          "description": "Key of a data structure to move from `.obsm` to `.obs`.",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        }
+      }
+    },
+    "mudata output": {
+      "title": "MuData Output",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/mudata input"
+    },
+    {
+      "$ref": "#/$defs/mudata output"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/metadata/move_obsm_to_obs/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/.config.vsh.yaml
new file mode 100644
index 0000000..f56986a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/.config.vsh.yaml
@@ -0,0 +1,386 @@
+name: "find_neighbors"
+namespace: "neighbors"
+version: "v4.0.4"
+authors:
+- name: "Dries De Maeyer"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "contributor"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_input"
+    description: "Which .obsm slot to use as a starting PCA embedding."
+    info: null
+    default:
+    - "X_pca"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file containing the found neighbors."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--uns_output"
+    description: "Mandatory .uns slot to store various neighbor output objects."
+    info: null
+    default:
+    - "neighbors"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsp_distances"
+    description: "In which .obsp slot to store the distance matrix between the resulting\
+      \ neighbors."
+    info: null
+    default:
+    - "distances"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsp_connectivities"
+    description: "In which .obsp slot to store the connectivities matrix between the\
+      \ resulting neighbors."
+    info: null
+    default:
+    - "connectivities"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--metric"
+    description: "The distance metric to be used in the generation of the nearest\
+      \ neighborhood network."
+    info: null
+    default:
+    - "euclidean"
+    required: false
+    choices:
+    - "cityblock"
+    - "cosine"
+    - "euclidean"
+    - "l1"
+    - "l2"
+    - "manhattan"
+    - "braycurtis"
+    - "canberra"
+    - "chebyshev"
+    - "correlation"
+    - "dice"
+    - "hamming"
+    - "jaccard"
+    - "kulsinski"
+    - "mahalanobis"
+    - "minkowski"
+    - "rogerstanimoto"
+    - "russellrao"
+    - "seuclidean"
+    - "sokalmichener"
+    - "sokalsneath"
+    - "sqeuclidean"
+    - "yule"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--num_neighbors"
+    description: "The size of local neighborhood (in terms of number of neighboring\
+      \ data points) used for manifold approximation. Larger values result in more\
+      \ global views of the manifold, while smaller values result in more local data\
+      \ being preserved. In general values should be in the range 2 to 100. If knn\
+      \ is True, number of nearest neighbors to be searched. If knn is False, a Gaussian\
+      \ kernel width is set to the distance of the n_neighbors neighbor."
+    info: null
+    default:
+    - 15
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--seed"
+    description: "A random seed."
+    info: null
+    default:
+    - 0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Compute a neighborhood graph of observations [McInnes18].\n\nThe neighbor\
+  \ search efficiency of this heavily relies on UMAP [McInnes18], which also provides\
+  \ a method for estimating connectivities of data points - the connectivity of the\
+  \ manifold (method=='umap'). If method=='gauss', connectivities are computed according\
+  \ to [Coifman05], in the adaption of [Haghverdi16].\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "lowcpu"
+    - "midmem"
+    - "middisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/neighbors/find_neighbors/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/neighbors/find_neighbors"
+  executable: "target/nextflow/neighbors/find_neighbors/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/main.nf
new file mode 100644
index 0000000..0fdbeef
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/main.nf
@@ -0,0 +1,4110 @@
+// find_neighbors v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries De Maeyer (maintainer)
+//  * Robrecht Cannoodt (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "find_neighbors",
+  "namespace" : "neighbors",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_input",
+          "description" : "Which .obsm slot to use as a starting PCA embedding.",
+          "default" : [
+            "X_pca"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file containing the found neighbors.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--uns_output",
+          "description" : "Mandatory .uns slot to store various neighbor output objects.",
+          "default" : [
+            "neighbors"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsp_distances",
+          "description" : "In which .obsp slot to store the distance matrix between the resulting neighbors.",
+          "default" : [
+            "distances"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsp_connectivities",
+          "description" : "In which .obsp slot to store the connectivities matrix between the resulting neighbors.",
+          "default" : [
+            "connectivities"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--metric",
+          "description" : "The distance metric to be used in the generation of the nearest neighborhood network.",
+          "default" : [
+            "euclidean"
+          ],
+          "required" : false,
+          "choices" : [
+            "cityblock",
+            "cosine",
+            "euclidean",
+            "l1",
+            "l2",
+            "manhattan",
+            "braycurtis",
+            "canberra",
+            "chebyshev",
+            "correlation",
+            "dice",
+            "hamming",
+            "jaccard",
+            "kulsinski",
+            "mahalanobis",
+            "minkowski",
+            "rogerstanimoto",
+            "russellrao",
+            "seuclidean",
+            "sokalmichener",
+            "sokalsneath",
+            "sqeuclidean",
+            "yule"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--num_neighbors",
+          "description" : "The size of local neighborhood (in terms of number of neighboring data points) used for manifold approximation. Larger values result in more global views of the manifold, while smaller values result in more local data being preserved. In general values should be in the range 2 to 100. If knn is True, number of nearest neighbors to be searched. If knn is False, a Gaussian kernel width is set to the distance of the n_neighbors neighbor.",
+          "default" : [
+            15
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--seed",
+          "description" : "A random seed.",
+          "default" : [
+            0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Compute a neighborhood graph of observations [McInnes18].\n\nThe neighbor search efficiency of this heavily relies on UMAP [McInnes18], which also provides a method for estimating connectivities of data points - the connectivity of the manifold (method=='umap'). If method=='gauss', connectivities are computed according to [Coifman05], in the adaption of [Haghverdi16].\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "lowcpu",
+          "midmem",
+          "middisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/neighbors/find_neighbors/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/neighbors/find_neighbors",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+import numpy as np
+
+numpy_module = sys.modules["numpy"]
+numpy_module.float_ = np.float64
+sys.modules["numpy"] = numpy_module
+
+import mudata as mu
+import scanpy as sc
+import sys
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsm_input': $( if [ ! -z ${VIASH_PAR_OBSM_INPUT+x} ]; then echo "r'${VIASH_PAR_OBSM_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'uns_output': $( if [ ! -z ${VIASH_PAR_UNS_OUTPUT+x} ]; then echo "r'${VIASH_PAR_UNS_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsp_distances': $( if [ ! -z ${VIASH_PAR_OBSP_DISTANCES+x} ]; then echo "r'${VIASH_PAR_OBSP_DISTANCES//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obsp_connectivities': $( if [ ! -z ${VIASH_PAR_OBSP_CONNECTIVITIES+x} ]; then echo "r'${VIASH_PAR_OBSP_CONNECTIVITIES//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'metric': $( if [ ! -z ${VIASH_PAR_METRIC+x} ]; then echo "r'${VIASH_PAR_METRIC//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'num_neighbors': $( if [ ! -z ${VIASH_PAR_NUM_NEIGHBORS+x} ]; then echo "int(r'${VIASH_PAR_NUM_NEIGHBORS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'seed': $( if [ ! -z ${VIASH_PAR_SEED+x} ]; then echo "int(r'${VIASH_PAR_SEED//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+logger.info("Reading input modality %s from %s", par["modality"], par["input"])
+adata = mu.read_h5ad(par["input"], mod=par["modality"])
+
+logger.info("Computing a neighborhood graph.")
+neighbors = sc.Neighbors(adata)
+neighbors.compute_neighbors(
+    n_neighbors=par["num_neighbors"],
+    use_rep=par["obsm_input"],
+    metric=par["metric"],
+    random_state=par["seed"],
+    method="umap",
+)
+
+adata.uns[par["uns_output"]] = {
+    "connectivities_key": par["obsp_connectivities"],
+    "distances_key": par["obsp_distances"],
+    "params": {
+        "n_neighbors": neighbors.n_neighbors,
+        "method": "umap",
+        "random_state": par["seed"],
+        "metric": par["metric"],
+        "use_rep": par["obsm_input"],
+    },
+}
+
+adata.obsp[par["obsp_distances"]] = neighbors.distances
+adata.obsp[par["obsp_connectivities"]] = neighbors.connectivities
+
+
+logger.info("Writing to %s", par["output"])
+write_h5ad_to_h5mu_with_compression(
+    par["output"], par["input"], par["modality"], adata, par["output_compression"]
+)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/neighbors/find_neighbors",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "lowcpu",
+    "midmem",
+    "middisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/nextflow.config
new file mode 100644
index 0000000..28023e5
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'neighbors/find_neighbors'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Compute a neighborhood graph of observations [McInnes18].\n\nThe neighbor search efficiency of this heavily relies on UMAP [McInnes18], which also provides a method for estimating connectivities of data points - the connectivity of the manifold (method==\'umap\'). If method==\'gauss\', connectivities are computed according to [Coifman05], in the adaption of [Haghverdi16].\n'
+  author = 'Dries De Maeyer, Robrecht Cannoodt'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/nextflow_schema.json
new file mode 100644
index 0000000..1f81de2
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/nextflow_schema.json
@@ -0,0 +1,131 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "find_neighbors",
+  "description": "Compute a neighborhood graph of observations [McInnes18].\n\nThe neighbor search efficiency of this heavily relies on UMAP [McInnes18], which also provides a method for estimating connectivities of data points - the connectivity of the manifold (method=='umap'). If method=='gauss', connectivities are computed according to [Coifman05], in the adaption of [Haghverdi16].\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "obsm_input": {
+          "type": "string",
+          "description": "Which .obsm slot to use as a starting PCA embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_pca\"`. ",
+          "default": "X_pca"
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file containing the found neighbors.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "uns_output": {
+          "type": "string",
+          "description": "Mandatory .uns slot to store various neighbor output objects.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"neighbors\"`. ",
+          "default": "neighbors"
+        },
+        "obsp_distances": {
+          "type": "string",
+          "description": "In which .obsp slot to store the distance matrix between the resulting neighbors.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"distances\"`. ",
+          "default": "distances"
+        },
+        "obsp_connectivities": {
+          "type": "string",
+          "description": "In which .obsp slot to store the connectivities matrix between the resulting neighbors.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"connectivities\"`. ",
+          "default": "connectivities"
+        },
+        "metric": {
+          "type": "string",
+          "description": "The distance metric to be used in the generation of the nearest neighborhood network.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"euclidean\"`, choices: ``cityblock`, `cosine`, `euclidean`, `l1`, `l2`, `manhattan`, `braycurtis`, `canberra`, `chebyshev`, `correlation`, `dice`, `hamming`, `jaccard`, `kulsinski`, `mahalanobis`, `minkowski`, `rogerstanimoto`, `russellrao`, `seuclidean`, `sokalmichener`, `sokalsneath`, `sqeuclidean`, `yule``. ",
+          "enum": [
+            "cityblock",
+            "cosine",
+            "euclidean",
+            "l1",
+            "l2",
+            "manhattan",
+            "braycurtis",
+            "canberra",
+            "chebyshev",
+            "correlation",
+            "dice",
+            "hamming",
+            "jaccard",
+            "kulsinski",
+            "mahalanobis",
+            "minkowski",
+            "rogerstanimoto",
+            "russellrao",
+            "seuclidean",
+            "sokalmichener",
+            "sokalsneath",
+            "sqeuclidean",
+            "yule"
+          ],
+          "default": "euclidean"
+        },
+        "num_neighbors": {
+          "type": "integer",
+          "description": "The size of local neighborhood (in terms of number of neighboring data points) used for manifold approximation",
+          "help_text": "Type: `integer`, multiple: `False`, default: `15`. ",
+          "default": 15
+        },
+        "seed": {
+          "type": "integer",
+          "description": "A random seed.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `0`. ",
+          "default": 0
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/neighbors/find_neighbors/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/.config.vsh.yaml
new file mode 100644
index 0000000..05e6db1
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/.config.vsh.yaml
@@ -0,0 +1,390 @@
+name: "calculate_qc_metrics"
+namespace: "qc"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process. \n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Layer from modality to use as input data. If not provided the .X\
+      \ attribute is used.\n"
+    info: null
+    example:
+    - "raw_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Metrics added to .obs"
+  arguments:
+  - type: "string"
+    name: "--var_qc_metrics"
+    description: "Keys to select a boolean (containing only True or False) column\
+      \ from .var.\nFor each cell, calculate the proportion of total values for genes\
+      \ which are labeled 'True', \ncompared to the total sum of the values for all\
+      \ genes.\n"
+    info: null
+    example:
+    - "ercc,highly_variable,mitochondrial"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--var_qc_metrics_fill_na_value"
+    description: "Fill any 'NA' values found in the columns specified with --var_qc_metrics\
+      \ to 'True' or 'False'.\nas False.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--top_n_vars"
+    description: "Number of top vars to be used to calculate cumulative proportions.\n\
+      If not specified, proportions are not calculated. `--top_n_vars 20;50` finds\n\
+      cumulative proportion to the 20th and 50th most expressed vars.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_num_nonzero_vars"
+    description: "Name of column in .obs describing, for each observation, the number\
+      \ of stored values\n(including explicit zeroes). In other words, the name of\
+      \ the column that counts\nfor each row the number of columns that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_vars"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_total_counts_vars"
+    description: "Name of the column for .obs describing, for each observation (row),\n\
+      the sum of the stored values in the columns.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Metrics added to .var"
+  arguments:
+  - type: "string"
+    name: "--output_var_num_nonzero_obs"
+    description: "Name of column describing, for each feature, the number of stored\
+      \ values\n(including explicit zeroes). In other words, the name of the column\
+      \ that counts\nfor each column the number of rows that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_obs"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_total_counts_obs"
+    description: "Name of the column in .var describing, for each feature (column),\n\
+      the sum of the stored values in the rows.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_obs_mean"
+    description: "Name of the column in .obs providing the mean of the values in each\
+      \ row.\n"
+    info: null
+    default:
+    - "obs_mean"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_pct_dropout"
+    description: "Name of the column in .obs providing for each feature the percentage\
+      \ of\nobservations the feature does not appear on (i.e. is missing). Same as\
+      \ `--num_nonzero_obs`\nbut percentage based.\n"
+    info: null
+    default:
+    - "pct_dropout"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Output h5mu file."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Add basic quality control metrics to an .h5mu file.\n\nThe metrics are\
+  \ comparable to what scanpy.pp.calculate_qc_metrics output,\nalthough they have\
+  \ slightly different names:\n\nVar metrics (name in this component -> name in scanpy):\n\
+  \  - pct_dropout -> pct_dropout_by_{expr_type}\n  - num_nonzero_obs -> n_cells_by_{expr_type}\n\
+  \  - obs_mean -> mean_{expr_type}\n  - total_counts -> total_{expr_type}\n\n  Obs\
+  \ metrics:\n    - num_nonzero_vars -> n_genes_by_{expr_type}\n    - pct_{var_qc_metrics}\
+  \ -> pct_{expr_type}_{qc_var}\n    - total_counts_{var_qc_metrics} -> total_{expr_type}_{qc_var}\n\
+  \    - pct_of_counts_in_top_{top_n_vars}_vars -> pct_{expr_type}_in_top_{n}_{var_type}\n\
+  \    - total_counts -> total_{expr_type}\n    \n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "singlecpu"
+    - "midmem"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.11-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scipy"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  - type: "python"
+    user: false
+    packages:
+    - "scanpy"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/qc/calculate_qc_metrics/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/qc/calculate_qc_metrics"
+  executable: "target/nextflow/qc/calculate_qc_metrics/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/main.nf
new file mode 100644
index 0000000..580ff6c
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/main.nf
@@ -0,0 +1,4233 @@
+// calculate_qc_metrics v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "calculate_qc_metrics",
+  "namespace" : "qc",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process. \n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Layer from modality to use as input data. If not provided the .X attribute is used.\n",
+          "example" : [
+            "raw_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Metrics added to .obs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_qc_metrics",
+          "description" : "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes.\n",
+          "example" : [
+            "ercc,highly_variable,mitochondrial"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--var_qc_metrics_fill_na_value",
+          "description" : "Fill any 'NA' values found in the columns specified with --var_qc_metrics to 'True' or 'False'.\nas False.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--top_n_vars",
+          "description" : "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated. `--top_n_vars 20;50` finds\ncumulative proportion to the 20th and 50th most expressed vars.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_num_nonzero_vars",
+          "description" : "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each row the number of columns that contain data.\n",
+          "default" : [
+            "num_nonzero_vars"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_total_counts_vars",
+          "description" : "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Metrics added to .var",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--output_var_num_nonzero_obs",
+          "description" : "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each column the number of rows that contain data.\n",
+          "default" : [
+            "num_nonzero_obs"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_total_counts_obs",
+          "description" : "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_obs_mean",
+          "description" : "Name of the column in .obs providing the mean of the values in each row.\n",
+          "default" : [
+            "obs_mean"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_pct_dropout",
+          "description" : "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e. is missing). Same as `--num_nonzero_obs`\nbut percentage based.\n",
+          "default" : [
+            "pct_dropout"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Output h5mu file.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Add basic quality control metrics to an .h5mu file.\n\nThe metrics are comparable to what scanpy.pp.calculate_qc_metrics output,\nalthough they have slightly different names:\n\nVar metrics (name in this component -> name in scanpy):\n  - pct_dropout -> pct_dropout_by_{expr_type}\n  - num_nonzero_obs -> n_cells_by_{expr_type}\n  - obs_mean -> mean_{expr_type}\n  - total_counts -> total_{expr_type}\n\n  Obs metrics:\n    - num_nonzero_vars -> n_genes_by_{expr_type}\n    - pct_{var_qc_metrics} -> pct_{expr_type}_{qc_var}\n    - total_counts_{var_qc_metrics} -> total_{expr_type}_{qc_var}\n    - pct_of_counts_in_top_{top_n_vars}_vars -> pct_{expr_type}_in_top_{n}_{var_type}\n    - total_counts -> total_{expr_type}\n    \n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_filtered_feature_bc_matrix.h5mu"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "singlecpu",
+          "midmem"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.11-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scipy"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "scanpy"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/qc/calculate_qc_metrics/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/qc/calculate_qc_metrics",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+from mudata import read_h5ad
+from scipy.sparse import issparse, csr_array
+import numpy as np
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'layer': $( if [ ! -z ${VIASH_PAR_LAYER+x} ]; then echo "r'${VIASH_PAR_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'var_qc_metrics': $( if [ ! -z ${VIASH_PAR_VAR_QC_METRICS+x} ]; then echo "r'${VIASH_PAR_VAR_QC_METRICS//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'var_qc_metrics_fill_na_value': $( if [ ! -z ${VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE+x} ]; then echo "r'${VIASH_PAR_VAR_QC_METRICS_FILL_NA_VALUE//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'top_n_vars': $( if [ ! -z ${VIASH_PAR_TOP_N_VARS+x} ]; then echo "list(map(int, r'${VIASH_PAR_TOP_N_VARS//\\'/\\'\\"\\'\\"r\\'}'.split(';')))"; else echo None; fi ),
+  'output_obs_num_nonzero_vars': $( if [ ! -z ${VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_OBS_NUM_NONZERO_VARS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_obs_total_counts_vars': $( if [ ! -z ${VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_OBS_TOTAL_COUNTS_VARS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_var_num_nonzero_obs': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_NUM_NONZERO_OBS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_var_total_counts_obs': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_TOTAL_COUNTS_OBS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_var_obs_mean': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_OBS_MEAN+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_OBS_MEAN//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_var_pct_dropout': $( if [ ! -z ${VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT_VAR_PCT_DROPOUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+
+def count_nonzero(layer, axis):
+    """
+    This method is the functional equivalent of the old .getnnz function from scirpy,
+    but that function was deprecated. So we use the nonzero function to mimic the old
+    behavior.
+    """
+    axis ^= 1
+    nonzero_counts = dict(zip(*np.unique(layer.nonzero()[axis], return_counts=True)))
+    nonzero_per_axis_item = {
+        row_index: nonzero_counts.get(row_index, 0)
+        for row_index in range(layer.shape[axis])
+    }
+    return np.array(list(nonzero_per_axis_item.values()), dtype="int64")
+
+
+def main():
+    modality_data = read_h5ad(par["input"], mod=par["modality"])
+    var = modality_data.var
+    layer = modality_data.X if not par["layer"] else modality_data.layers[par["layer"]]
+    if not issparse(layer):
+        raise NotImplementedError("Expected layer to be in sparse format.")
+    layer = csr_array(layer)
+    layer.eliminate_zeros()
+
+    var_columns_to_add = {}
+
+    # var statistics
+    if par["output_var_obs_mean"]:
+        obs_mean = layer.mean(axis=0)
+        var_columns_to_add[par["output_var_obs_mean"]] = obs_mean
+    if par["output_var_total_counts_obs"]:
+        # from the np.sum documentation:
+        # Especially when summing a large number of lower precision floating point numbers,
+        # such as float32, numerical errors can become significant. In such cases it can
+        # be advisable to use dtype="float64" to use a higher precision for the output.
+        layer_with_type = layer
+        if np.issubdtype(layer.dtype, np.floating) and np.can_cast(
+            layer.dtype, np.float64, casting="safe"
+        ):
+            # 'safe' casting makes sure not to cast np.float128 or anything else to a lower precision dtype
+            layer_with_type = layer.astype(np.float64)
+        total_counts_obs = np.ravel(layer_with_type.sum(axis=0))
+        var_columns_to_add[par["output_var_total_counts_obs"]] = total_counts_obs
+
+    num_nonzero_obs = count_nonzero(layer, axis=0)
+    if par["output_var_num_nonzero_obs"]:
+        var_columns_to_add[par["output_var_num_nonzero_obs"]] = num_nonzero_obs
+    if par["output_var_pct_dropout"]:
+        var_columns_to_add[par["output_var_pct_dropout"]] = (
+            1 - num_nonzero_obs / layer.shape[0]
+        ) * 100
+
+    modality_data.var = modality_data.var.assign(**var_columns_to_add)
+
+    # obs statistics
+    obs_columns_to_add = {}
+    total_counts_var = np.ravel(layer.sum(axis=1))
+
+    if par["output_obs_num_nonzero_vars"]:
+        num_nonzero_vars = count_nonzero(layer, axis=1)
+        obs_columns_to_add[par["output_obs_num_nonzero_vars"]] = num_nonzero_vars
+
+    if par["output_obs_total_counts_vars"]:
+        obs_columns_to_add[par["output_obs_total_counts_vars"]] = total_counts_var
+
+    top_metrics = {}
+    if par["top_n_vars"]:
+        par["top_n_vars"] = sorted(par["top_n_vars"])
+        distributions = get_top_from_csr_matrix(layer, par["top_n_vars"])
+        top_metrics = {
+            distribution_size: distribution * 100
+            for distribution_size, distribution in zip(
+                par["top_n_vars"], distributions.T
+            )
+        }
+        obs_columns_to_add |= {
+            f"pct_of_counts_in_top_{n_top}_vars": col
+            for n_top, col in top_metrics.items()
+        }
+
+    if par["var_qc_metrics"]:
+        print(f"qc_metrics: {par['var_qc_metrics']}")
+        for qc_metric in par["var_qc_metrics"]:
+            if qc_metric not in var:
+                raise ValueError(
+                    f"Value for --var_qc_metrics, '{qc_metric}' "
+                    f"not found in .var for modality {par['modality']}"
+                )
+            qc_column = var[qc_metric]
+            if qc_column.isna().any():
+                if par["var_qc_metrics_fill_na_value"] is None:
+                    raise ValueError(
+                        f"The .var column '{qc_metric}', selected by '--var_qc_metrics', contains NA values. "
+                        "It is ambiguous whether or not to include these values in the static calulation. "
+                        "You can explicitly map the NA values to 'False' or 'True using '--var_qc_metrics_fill_na_value'"
+                    )
+                else:
+                    qc_column = qc_column.fillna(
+                        par["var_qc_metrics_fill_na_value"], inplace=False
+                    )
+            qc_column = qc_column.to_list()
+            if set(np.unique(qc_column)) - {True, False}:
+                raise ValueError(
+                    f"Column {qc_metric} in .var for modality {par['modality']} "
+                    f"must only contain boolean values"
+                )
+            total_counts_qc_metric = np.ravel(layer[:, qc_column].sum(axis=1))
+            obs_columns_to_add |= {
+                f"total_counts_{qc_metric}": total_counts_qc_metric,
+                f"pct_{qc_metric}": total_counts_qc_metric / total_counts_var * 100,
+            }
+
+    modality_data.obs = modality_data.obs.assign(**obs_columns_to_add)
+    write_h5ad_to_h5mu_with_compression(
+        par["output"],
+        par["input"],
+        par["modality"],
+        modality_data,
+        par["output_compression"],
+    )
+
+
+def get_top_from_csr_matrix(array, top_n_genes):
+    # csr matrices stores a 3D matrix in a format such that data for individual cells
+    # are stored in 1 array. Another array (indptr) here stores the ranges of indices
+    # to select from the data-array (.e.g. data[indptr[0]:indptr[1]] for row 0) for each row.
+    # Another array 'indices' maps each element of data to a column
+    # (data and indices arrays have the same length)
+    top_n_genes = np.array(top_n_genes).astype(np.int64)
+    assert np.all(top_n_genes[:-1] <= top_n_genes[1:]), "top_n_genes must be sorted"
+    row_indices, data = array.indptr, array.data
+    number_of_rows, max_genes_to_parse = row_indices.size - 1, top_n_genes[-1]
+    top_data = np.zeros((number_of_rows, max_genes_to_parse), dtype=data.dtype)
+    # Loop over each row to create a dense matrix without the 0 counts,
+    # but not for the whole matrix, only store the genes up until
+    # the largest number of top n genes.
+    for row_number in range(number_of_rows):
+        row_start_index, row_end_index = (
+            row_indices[row_number],
+            row_indices[row_number + 1],
+        )
+        row_data = data[row_start_index:row_end_index]  # all non-zero counts for an row
+        try:
+            # There are less genes with counts in the row than the
+            # maximum number of genes we would like to select
+            # all these genes are in the top genes, just store them
+            top_data[row_number, : row_end_index - row_start_index] = row_data
+        except ValueError:
+            # Store the counts for the top genes
+            top_data[row_number, :] = np.partition(row_data, -max_genes_to_parse)[
+                -max_genes_to_parse:
+            ]
+
+    # Partition works from smallest to largest, but we want largest
+    # so do smallest to largest first (but with reversed indices)
+    top_data = np.partition(top_data, max_genes_to_parse - top_n_genes)
+    # And then switch the order around
+    top_data = np.flip(top_data, axis=1)
+
+    cumulative = top_data.cumsum(axis=1, dtype=np.float64)[:, top_n_genes - 1]
+    return cumulative / np.expand_dims(array.sum(axis=1), 1)
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/qc/calculate_qc_metrics",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "singlecpu",
+    "midmem"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/nextflow.config
new file mode 100644
index 0000000..07dd350
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'qc/calculate_qc_metrics'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Add basic quality control metrics to an .h5mu file.\n\nThe metrics are comparable to what scanpy.pp.calculate_qc_metrics output,\nalthough they have slightly different names:\n\nVar metrics (name in this component -> name in scanpy):\n  - pct_dropout -> pct_dropout_by_{expr_type}\n  - num_nonzero_obs -> n_cells_by_{expr_type}\n  - obs_mean -> mean_{expr_type}\n  - total_counts -> total_{expr_type}\n\n  Obs metrics:\n    - num_nonzero_vars -> n_genes_by_{expr_type}\n    - pct_{var_qc_metrics} -> pct_{expr_type}_{qc_var}\n    - total_counts_{var_qc_metrics} -> total_{expr_type}_{qc_var}\n    - pct_of_counts_in_top_{top_n_vars}_vars -> pct_{expr_type}_in_top_{n}_{var_type}\n    - total_counts -> total_{expr_type}\n    \n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/nextflow_schema.json
new file mode 100644
index 0000000..83f9504
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/nextflow_schema.json
@@ -0,0 +1,156 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "calculate_qc_metrics",
+  "description": "Add basic quality control metrics to an .h5mu file.\n\nThe metrics are comparable to what scanpy.pp.calculate_qc_metrics output,\nalthough they have slightly different names:\n\nVar metrics (name in this component -> name in scanpy):\n  - pct_dropout -> pct_dropout_by_{expr_type}\n  - num_nonzero_obs -> n_cells_by_{expr_type}\n  - obs_mean -> mean_{expr_type}\n  - total_counts -> total_{expr_type}\n\n  Obs metrics:\n    - num_nonzero_vars -> n_genes_by_{expr_type}\n    - pct_{var_qc_metrics} -> pct_{expr_type}_{qc_var}\n    - total_counts_{var_qc_metrics} -> total_{expr_type}_{qc_var}\n    - pct_of_counts_in_top_{top_n_vars}_vars -> pct_{expr_type}_in_top_{n}_{var_type}\n    - total_counts -> total_{expr_type}\n    \n",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "layer": {
+          "type": "string",
+          "description": "Layer from modality to use as input data",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"raw_counts\"`. "
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "metrics added to .obs": {
+      "title": "Metrics added to .obs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_qc_metrics": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes.\n",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"ercc,highly_variable,mitochondrial\"]`. "
+        },
+        "var_qc_metrics_fill_na_value": {
+          "type": "boolean",
+          "description": "Fill any 'NA' values found in the columns specified with --var_qc_metrics to 'True' or 'False'.\nas False.\n",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        },
+        "top_n_vars": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated",
+          "help_text": "Type: `integer`, multiple: `True`. "
+        },
+        "output_obs_num_nonzero_vars": {
+          "type": "string",
+          "description": "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes)",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"num_nonzero_vars\"`. ",
+          "default": "num_nonzero_vars"
+        },
+        "output_obs_total_counts_vars": {
+          "type": "string",
+          "description": "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"total_counts\"`. ",
+          "default": "total_counts"
+        }
+      }
+    },
+    "metrics added to .var": {
+      "title": "Metrics added to .var",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output_var_num_nonzero_obs": {
+          "type": "string",
+          "description": "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes)",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"num_nonzero_obs\"`. ",
+          "default": "num_nonzero_obs"
+        },
+        "output_var_total_counts_obs": {
+          "type": "string",
+          "description": "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"total_counts\"`. ",
+          "default": "total_counts"
+        },
+        "output_var_obs_mean": {
+          "type": "string",
+          "description": "Name of the column in .obs providing the mean of the values in each row.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"obs_mean\"`. ",
+          "default": "obs_mean"
+        },
+        "output_var_pct_dropout": {
+          "type": "string",
+          "description": "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"pct_dropout\"`. ",
+          "default": "pct_dropout"
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/metrics added to .obs"
+    },
+    {
+      "$ref": "#/$defs/metrics added to .var"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/qc/calculate_qc_metrics/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/.config.vsh.yaml
new file mode 100644
index 0000000..128756f
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/.config.vsh.yaml
@@ -0,0 +1,288 @@
+name: "clr"
+namespace: "transform"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "prot"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file."
+    info: null
+    default:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_layer"
+    description: "Input layer to use. By default, .X is used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_layer"
+    description: "Output layer to use. By default, use X."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--axis"
+    description: "Axis across which CLR is performed. If set to 0, CLR is performed\
+      \ across observations (cells).\nIf set to 1, CLR is performed across features\
+      \ (genes).\n"
+    info: null
+    default:
+    - 0
+    required: false
+    choices:
+    - 0
+    - 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Perform CLR normalization on CITE-seq data (Stoeckius et al., 2017).\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "lowmem"
+    - "midcpu"
+    - "middisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.13-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    - "muon~=0.1.7"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/transform/clr/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/transform/clr"
+  executable: "target/nextflow/transform/clr/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/main.nf
new file mode 100644
index 0000000..8c3fb63
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/main.nf
@@ -0,0 +1,3980 @@
+// clr v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "clr",
+  "namespace" : "transform",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "prot"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file.",
+          "default" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_layer",
+          "description" : "Input layer to use. By default, .X is used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_layer",
+          "description" : "Output layer to use. By default, use X.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--axis",
+          "description" : "Axis across which CLR is performed. If set to 0, CLR is performed across observations (cells).\nIf set to 1, CLR is performed across features (genes).\n",
+          "default" : [
+            0
+          ],
+          "required" : false,
+          "choices" : [
+            0,
+            1
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Perform CLR normalization on CITE-seq data (Stoeckius et al., 2017).\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "lowmem",
+          "midcpu",
+          "middisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.13-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4",
+            "muon~=0.1.7"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/transform/clr/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/transform/clr",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+from muon import prot as pt
+from mudata import read_h5ad
+from anndata import AnnData
+from functools import partial
+from operator import setitem
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_layer': $( if [ ! -z ${VIASH_PAR_INPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_INPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_layer': $( if [ ! -z ${VIASH_PAR_OUTPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_OUTPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'axis': $( if [ ! -z ${VIASH_PAR_AXIS+x} ]; then echo "int(r'${VIASH_PAR_AXIS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+
+sys.path.append(meta["resources_dir"])
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+
+def main():
+    input_data = read_h5ad(par["input"], mod=par["modality"])
+    if par["input_layer"]:
+        input_data = AnnData(X=input_data.layers[par["input_layer"]])
+    # CLR always normalizes the .X layer, so we have to create an AnnData file with
+    # the input layer at .X
+    normalized_counts = pt.pp.clr(input_data, axis=par["axis"], inplace=False)
+    if not normalized_counts:
+        raise RuntimeError("CLR failed to return the requested output layer")
+
+    output_layer_setter = (
+        partial(setattr, input_data, "X")
+        if not par["output_layer"]
+        else partial(setitem, input_data.layers, par["output_layer"])
+    )
+    output_layer_setter(normalized_counts.X)
+    write_h5ad_to_h5mu_with_compression(
+        par["output"],
+        par["input"],
+        par["modality"],
+        input_data,
+        par["output_compression"],
+    )
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/transform/clr",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "lowmem",
+    "midcpu",
+    "middisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/nextflow.config
new file mode 100644
index 0000000..2cf27bc
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'transform/clr'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Perform CLR normalization on CITE-seq data (Stoeckius et al., 2017).\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/nextflow_schema.json
new file mode 100644
index 0000000..cf13db6
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/clr/nextflow_schema.json
@@ -0,0 +1,84 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "clr",
+  "description": "Perform CLR normalization on CITE-seq data (Stoeckius et al., 2017).\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"prot\"`. ",
+          "default": "prot"
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"output.h5mu\"`, direction: `output`. ",
+          "default": "output.h5mu"
+        },
+        "input_layer": {
+          "type": "string",
+          "description": "Input layer to use",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "output_layer": {
+          "type": "string",
+          "description": "Output layer to use",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "axis": {
+          "type": "integer",
+          "description": "Axis across which CLR is performed",
+          "help_text": "Type: `integer`, multiple: `False`, default: `0`, choices: ``0`, `1``. ",
+          "enum": [
+            0,
+            1
+          ],
+          "default": 0
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/.config.vsh.yaml
new file mode 100644
index 0000000..123e551
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/.config.vsh.yaml
@@ -0,0 +1,275 @@
+name: "delete_layer"
+namespace: "transform"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Input layer to remove"
+    info: null
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file."
+    info: null
+    default:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--missing_ok"
+    description: "Do not raise an error if the layer does not exist for all modalities."
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Delete an anndata layer from one or more modalities.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "midmem"
+    - "singlecpu"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/transform/delete_layer/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/transform/delete_layer"
+  executable: "target/nextflow/transform/delete_layer/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/main.nf
new file mode 100644
index 0000000..a50c75f
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/main.nf
@@ -0,0 +1,3970 @@
+// delete_layer v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "delete_layer",
+  "namespace" : "transform",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Input layer to remove",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file.",
+          "default" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--missing_ok",
+          "description" : "Do not raise an error if the layer does not exist for all modalities.",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Delete an anndata layer from one or more modalities.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "midmem",
+          "singlecpu"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/transform/delete_layer/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/transform/delete_layer",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+from mudata import read_h5ad
+from pathlib import Path
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'layer': $( if [ ! -z ${VIASH_PAR_LAYER+x} ]; then echo "r'${VIASH_PAR_LAYER//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'missing_ok': $( if [ ! -z ${VIASH_PAR_MISSING_OK+x} ]; then echo "r'${VIASH_PAR_MISSING_OK//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+
+def main():
+    input_file, output_file, mod_name = (
+        Path(par["input"]),
+        Path(par["output"]),
+        par["modality"],
+    )
+
+    logger.info("Reading input file %s, modality %s.", input_file, mod_name)
+    mod = read_h5ad(input_file, mod=mod_name)
+    for layer in par["layer"]:
+        if layer not in mod.layers:
+            if par["missing_ok"]:
+                continue
+            raise ValueError(f"Layer '{layer}' is not present in modality {mod_name}.")
+        logger.info("Deleting layer %s from modality %s.", layer, mod_name)
+        del mod.layers[layer]
+
+    logger.info("Writing output to %s.", par["output"])
+
+    write_h5ad_to_h5mu_with_compression(
+        output_file, input_file, mod_name, mod, par["output_compression"]
+    )
+
+    logger.info("Finished.")
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/transform/delete_layer",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "midmem",
+    "singlecpu"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/nextflow.config
new file mode 100644
index 0000000..156248d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'transform/delete_layer'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Delete an anndata layer from one or more modalities.\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/nextflow_schema.json
new file mode 100644
index 0000000..c2a74fc
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/nextflow_schema.json
@@ -0,0 +1,78 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "delete_layer",
+  "description": "Delete an anndata layer from one or more modalities.\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "layer": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Input layer to remove",
+          "help_text": "Type: `string`, multiple: `True`, required. "
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"output.h5mu\"`, direction: `output`. ",
+          "default": "output.h5mu"
+        },
+        "missing_ok": {
+          "type": "boolean",
+          "description": "Do not raise an error if the layer does not exist for all modalities.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/delete_layer/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/.config.vsh.yaml
new file mode 100644
index 0000000..0c85e2f
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/.config.vsh.yaml
@@ -0,0 +1,301 @@
+name: "log1p"
+namespace: "transform"
+version: "v4.0.4"
+authors:
+- name: "Dries De Maeyer"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "contributor"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_layer"
+    description: "Input layer to use. If None, X is normalized"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_layer"
+    description: "Output layer to use. By default, use X."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file."
+    info: null
+    default:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--base"
+    description: "Base of the logarithm. Natural logarithm is used by default.\n"
+    info: null
+    example:
+    - 2.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Logarithmize the data matrix. Computes X = log(X + 1), where log denotes\
+  \ the natural logarithm unless a different base is given.\n"
+test_resources:
+- type: "python_script"
+  path: "run_test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "midmem"
+    - "lowcpu"
+    - "highdisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/transform/log1p/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/transform/log1p"
+  executable: "target/nextflow/transform/log1p/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/main.nf
new file mode 100644
index 0000000..cfcbbbd
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/main.nf
@@ -0,0 +1,4006 @@
+// log1p v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries De Maeyer (maintainer)
+//  * Robrecht Cannoodt (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "log1p",
+  "namespace" : "transform",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_layer",
+          "description" : "Input layer to use. If None, X is normalized",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_layer",
+          "description" : "Output layer to use. By default, use X.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file.",
+          "default" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--base",
+          "description" : "Base of the logarithm. Natural logarithm is used by default.\n",
+          "example" : [
+            2.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Logarithmize the data matrix. Computes X = log(X + 1), where log denotes the natural logarithm unless a different base is given.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "run_test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "midmem",
+          "lowcpu",
+          "highdisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/transform/log1p/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/transform/log1p",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import scanpy as sc
+import mudata as mu
+import anndata as ad
+import sys
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_layer': $( if [ ! -z ${VIASH_PAR_INPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_INPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_layer': $( if [ ! -z ${VIASH_PAR_OUTPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_OUTPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'base': $( if [ ! -z ${VIASH_PAR_BASE+x} ]; then echo "float(r'${VIASH_PAR_BASE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+logger.info("Reading modality %s from input mudata", par["modality"])
+data = mu.read_h5ad(par["input"], mod=par["modality"])
+assert data.var_names.is_unique, "Expected var_names to be unique."
+
+logger.info("Performing log transformation")
+# Make our own copy with not a lot of data
+# this avoid excessive memory usage and accidental overwrites
+input_layer = data.layers[par["input_layer"]] if par["input_layer"] else data.X
+data_for_scanpy = ad.AnnData(X=input_layer.copy())
+sc.pp.log1p(
+    data_for_scanpy,
+    base=par["base"],
+    layer=None,  # use X
+    copy=False,
+)  # allow overwrites in the copy that was made
+
+# Scanpy will overwrite the input layer.
+# So fetch input layer from the copy and use it to populate the output slot
+if par["output_layer"]:
+    data.layers[par["output_layer"]] = data_for_scanpy.X
+else:
+    data.X = data_for_scanpy.X
+data.uns["log1p"] = data_for_scanpy.uns["log1p"].copy()
+
+logger.info("Writing to file %s", par["output"])
+write_h5ad_to_h5mu_with_compression(
+    par["output"], par["input"], par["modality"], data, par["output_compression"]
+)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/transform/log1p",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "midmem",
+    "lowcpu",
+    "highdisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/nextflow.config
new file mode 100644
index 0000000..e576563
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'transform/log1p'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Logarithmize the data matrix. Computes X = log(X + 1), where log denotes the natural logarithm unless a different base is given.\n'
+  author = 'Dries De Maeyer, Robrecht Cannoodt'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/nextflow_schema.json
new file mode 100644
index 0000000..86612bf
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/nextflow_schema.json
@@ -0,0 +1,79 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "log1p",
+  "description": "Logarithmize the data matrix. Computes X = log(X + 1), where log denotes the natural logarithm unless a different base is given.\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "input_layer": {
+          "type": "string",
+          "description": "Input layer to use",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "output_layer": {
+          "type": "string",
+          "description": "Output layer to use",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"output.h5mu\"`, direction: `output`. ",
+          "default": "output.h5mu"
+        },
+        "base": {
+          "type": "number",
+          "description": "Base of the logarithm",
+          "help_text": "Type: `double`, multiple: `False`, example: `2.0`. "
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/log1p/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/.config.vsh.yaml
new file mode 100644
index 0000000..7b898ef
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/.config.vsh.yaml
@@ -0,0 +1,319 @@
+name: "normalize_total"
+namespace: "transform"
+version: "v4.0.4"
+authors:
+- name: "Dries De Maeyer"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "contributor"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality from the input MuData file to process.\n"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_layer"
+    description: "Input layer to use. By default, X is normalized"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file."
+    info: null
+    default:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_layer"
+    description: "Output layer to use. By default, use X."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--target_sum"
+    description: "If None, after normalization, each observation (cell) has a total\
+      \ count equal to the median of total counts for observations (cells) before\
+      \ normalization."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--exclude_highly_expressed"
+    description: "Exclude (very) highly expressed genes for the computation of the\
+      \ normalization factor (size factor) for each cell. A gene is considered highly\
+      \ expressed, if it has more than max_fraction of the total counts in at least\
+      \ one cell. The not-excluded genes will sum up to target_sum."
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Normalize counts per cell.\n\nNormalize each cell by total counts over\
+  \ all genes, so that every cell has the same total count after normalization. If\
+  \ choosing target_sum=1e6, this is CPM normalization.\n\nIf exclude_highly_expressed=True,\
+  \ very highly expressed genes are excluded from the computation of the normalization\
+  \ factor (size factor) for each cell. This is meaningful as these can strongly influence\
+  \ the resulting normalized values for all other genes [Weinreb17].\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "midmem"
+    - "lowcpu"
+    - "highdisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/transform/normalize_total/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/transform/normalize_total"
+  executable: "target/nextflow/transform/normalize_total/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/main.nf
new file mode 100644
index 0000000..b90e7db
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/main.nf
@@ -0,0 +1,4023 @@
+// normalize_total v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries De Maeyer (maintainer)
+//  * Robrecht Cannoodt (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "normalize_total",
+  "namespace" : "transform",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality from the input MuData file to process.\n",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_layer",
+          "description" : "Input layer to use. By default, X is normalized",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file.",
+          "default" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_layer",
+          "description" : "Output layer to use. By default, use X.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--target_sum",
+          "description" : "If None, after normalization, each observation (cell) has a total count equal to the median of total counts for observations (cells) before normalization.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--exclude_highly_expressed",
+          "description" : "Exclude (very) highly expressed genes for the computation of the normalization factor (size factor) for each cell. A gene is considered highly expressed, if it has more than max_fraction of the total counts in at least one cell. The not-excluded genes will sum up to target_sum.",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Normalize counts per cell.\n\nNormalize each cell by total counts over all genes, so that every cell has the same total count after normalization. If choosing target_sum=1e6, this is CPM normalization.\n\nIf exclude_highly_expressed=True, very highly expressed genes are excluded from the computation of the normalization factor (size factor) for each cell. This is meaningful as these can strongly influence the resulting normalized values for all other genes [Weinreb17].\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "midmem",
+          "lowcpu",
+          "highdisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/transform/normalize_total/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/transform/normalize_total",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+import scanpy as sc
+import mudata as mu
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_layer': $( if [ ! -z ${VIASH_PAR_INPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_INPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_layer': $( if [ ! -z ${VIASH_PAR_OUTPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_OUTPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'target_sum': $( if [ ! -z ${VIASH_PAR_TARGET_SUM+x} ]; then echo "int(r'${VIASH_PAR_TARGET_SUM//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'exclude_highly_expressed': $( if [ ! -z ${VIASH_PAR_EXCLUDE_HIGHLY_EXPRESSED+x} ]; then echo "r'${VIASH_PAR_EXCLUDE_HIGHLY_EXPRESSED//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+logger.info("Reading modality %s from %s", par["modality"], par["input"])
+dat = mu.read_h5ad(par["input"], mod=par["modality"])
+assert dat.var_names.is_unique, "The var_names of the input modality must be be unique."
+
+logger.info(par)
+
+logger.info("Performing total normalization.")
+if par["input_layer"] and par["input_layer"] not in dat.layers.keys():
+    raise ValueError(f"Input layer {par['input_layer']} not found in {par['modality']}")
+output_data = sc.pp.normalize_total(
+    dat,
+    layer=par["input_layer"],
+    target_sum=par["target_sum"],
+    copy=True if par["output_layer"] else False,
+)
+
+if output_data:
+    result = (
+        output_data.X
+        if not par["input_layer"]
+        else output_data.layers[par["input_layer"]]
+    )
+    dat.layers[par["output_layer"]] = result
+
+logger.info(
+    "Writing to file to %s with compression %s",
+    par["output"],
+    par["output_compression"],
+)
+write_h5ad_to_h5mu_with_compression(
+    par["output"], par["input"], par["modality"], dat, par["output_compression"]
+)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/transform/normalize_total",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "midmem",
+    "lowcpu",
+    "highdisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/nextflow.config
new file mode 100644
index 0000000..afa360b
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'transform/normalize_total'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Normalize counts per cell.\n\nNormalize each cell by total counts over all genes, so that every cell has the same total count after normalization. If choosing target_sum=1e6, this is CPM normalization.\n\nIf exclude_highly_expressed=True, very highly expressed genes are excluded from the computation of the normalization factor (size factor) for each cell. This is meaningful as these can strongly influence the resulting normalized values for all other genes [Weinreb17].\n'
+  author = 'Dries De Maeyer, Robrecht Cannoodt'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/nextflow_schema.json
new file mode 100644
index 0000000..2592614
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/nextflow_schema.json
@@ -0,0 +1,85 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "normalize_total",
+  "description": "Normalize counts per cell.\n\nNormalize each cell by total counts over all genes, so that every cell has the same total count after normalization. If choosing target_sum=1e6, this is CPM normalization.\n\nIf exclude_highly_expressed=True, very highly expressed genes are excluded from the computation of the normalization factor (size factor) for each cell. This is meaningful as these can strongly influence the resulting normalized values for all other genes [Weinreb17].\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality from the input MuData file to process.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "input_layer": {
+          "type": "string",
+          "description": "Input layer to use",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"output.h5mu\"`, direction: `output`. ",
+          "default": "output.h5mu"
+        },
+        "output_layer": {
+          "type": "string",
+          "description": "Output layer to use",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "target_sum": {
+          "type": "integer",
+          "description": "If None, after normalization, each observation (cell) has a total count equal to the median of total counts for observations (cells) before normalization.",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "exclude_highly_expressed": {
+          "type": "boolean",
+          "description": "Exclude (very) highly expressed genes for the computation of the normalization factor (size factor) for each cell",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/normalize_total/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/.config.vsh.yaml
new file mode 100644
index 0000000..3a6ce9d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/.config.vsh.yaml
@@ -0,0 +1,294 @@
+name: "scale"
+namespace: "transform"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Arguments"
+  arguments:
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Input h5mu file."
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "List of modalities to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_layer"
+    description: "Input layer with data to scale. Uses .X by default"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_layer"
+    description: "Output layer where scaled data will be stored. If not specified,\
+      \ .X will be used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--max_value"
+    description: "Clip (truncate) to this value after scaling. Does not clip by default."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_false"
+    name: "--zero_center"
+    description: "If set, omit zero-centering variables, which allows to handle sparse\
+      \ input efficiently."
+    info: null
+    direction: "input"
+  - type: "file"
+    name: "--output"
+    alternatives:
+    - "-o"
+    description: "Output h5mu file."
+    info: null
+    default:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "Compression format to use for the output AnnData and/or Mudata objects.\n\
+      By default no compression is applied.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "compress_h5mu.py"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Scale data to unit variance and zero mean.\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "lowmem"
+    - "lowcpu"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v4.0.4"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.12.7"
+    - "awkward"
+    - "mudata~=0.3.2"
+    - "scanpy~=1.11.4"
+    script:
+    - "exec(\"try:\\n  import zarr; from importlib.metadata import version\\nexcept\
+      \ ModuleNotFoundError:\\n  exit(0)\\nelse:  assert int(version(\\\"zarr\\\"\
+      ).partition(\\\".\\\")[0]) > 2\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/transform/scale/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/transform/scale"
+  executable: "target/nextflow/transform/scale/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/compress_h5mu.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/compress_h5mu.py
new file mode 100644
index 0000000..4b363ee
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/compress_h5mu.py
@@ -0,0 +1,87 @@
+import shutil
+from anndata import AnnData
+from mudata import write_h5ad
+from h5py import File as H5File
+from h5py import Group, Dataset
+from pathlib import Path
+from typing import Union, Literal
+from functools import partial
+
+
+def compress_h5mu(
+    input_path: Union[str, Path],
+    output_path: Union[str, Path],
+    compression: Union[Literal["gzip"], Literal["lzf"]],
+):
+    input_path, output_path = str(input_path), str(output_path)
+
+    def copy_attributes(in_object, out_object):
+        for key, value in in_object.attrs.items():
+            out_object.attrs[key] = value
+
+    def visit_path(
+        output_h5: H5File,
+        compression: Union[Literal["gzip"], Literal["lzf"]],
+        name: str,
+        object: Union[Group, Dataset],
+    ):
+        if isinstance(object, Group):
+            new_group = output_h5.create_group(name)
+            copy_attributes(object, new_group)
+        elif isinstance(object, Dataset):
+            # Compression only works for non-scalar Dataset objects
+            # Scalar objects dont have a shape defined
+            if not object.compression and object.shape not in [None, ()]:
+                new_dataset = output_h5.create_dataset(
+                    name, data=object, compression=compression
+                )
+                copy_attributes(object, new_dataset)
+            else:
+                output_h5.copy(object, name)
+        else:
+            raise NotImplementedError(
+                f"Could not copy element {name}, "
+                f"type has not been implemented yet: {type(object)}"
+            )
+
+    with (
+        H5File(input_path, "r") as input_h5,
+        H5File(output_path, "w", userblock_size=512) as output_h5,
+    ):
+        copy_attributes(input_h5, output_h5)
+        input_h5.visititems(partial(visit_path, output_h5, compression))
+
+    with open(input_path, "rb") as input_bytes:
+        # Mudata puts metadata like this in the first 512 bytes:
+        # MuData (format-version=0.1.0;creator=muon;creator-version=0.2.0)
+        # See mudata/_core/io.py, read_h5mu() function
+        starting_metadata = input_bytes.read(100)
+        # The metadata is padded with extra null bytes up until 512 bytes
+        truncate_location = starting_metadata.find(b"\x00")
+        starting_metadata = starting_metadata[:truncate_location]
+    with open(output_path, "br+") as f:
+        nbytes = f.write(starting_metadata)
+        f.write(b"\0" * (512 - nbytes))
+
+
+def write_h5ad_to_h5mu_with_compression(
+    output_file: Union[str, Path],
+    h5mu: Union[str, Path],
+    modality_name: str,
+    modality_data: AnnData,
+    output_compression=None,
+):
+    output_file = Path(output_file)
+    h5mu = Path(h5mu)
+    output_file_uncompressed = (
+        output_file.with_name(output_file.stem + "_uncompressed.h5mu")
+        if output_compression
+        else output_file
+    )
+    shutil.copyfile(h5mu, output_file_uncompressed)
+    write_h5ad(filename=output_file_uncompressed, mod=modality_name, data=modality_data)
+    if output_compression:
+        compress_h5mu(
+            output_file_uncompressed, output_file, compression=output_compression
+        )
+        output_file_uncompressed.unlink()
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/main.nf
new file mode 100644
index 0000000..4965425
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/main.nf
@@ -0,0 +1,3999 @@
+// scale v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "scale",
+  "namespace" : "transform",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Input h5mu file.",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "List of modalities to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_layer",
+          "description" : "Input layer with data to scale. Uses .X by default",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_layer",
+          "description" : "Output layer where scaled data will be stored. If not specified, .X will be used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--max_value",
+          "description" : "Clip (truncate) to this value after scaling. Does not clip by default.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_false",
+          "name" : "--zero_center",
+          "description" : "If set, omit zero-centering variables, which allows to handle sparse input efficiently.",
+          "direction" : "input"
+        },
+        {
+          "type" : "file",
+          "name" : "--output",
+          "alternatives" : [
+            "-o"
+          ],
+          "description" : "Output h5mu file.",
+          "default" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/compress_h5mu.py"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Scale data to unit variance and zero mean.\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "lowmem",
+          "lowcpu"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v4.0.4",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.12.7",
+            "awkward",
+            "mudata~=0.3.2",
+            "scanpy~=1.11.4"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import zarr; from importlib.metadata import version\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  assert int(version(\\\\\\"zarr\\\\\\").partition(\\\\\\".\\\\\\")[0]) > 2\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/transform/scale/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/transform/scale",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import sys
+from mudata import read_h5ad
+import scanpy
+from functools import partial
+from operator import setitem
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'input_layer': $( if [ ! -z ${VIASH_PAR_INPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_INPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_layer': $( if [ ! -z ${VIASH_PAR_OUTPUT_LAYER+x} ]; then echo "r'${VIASH_PAR_OUTPUT_LAYER//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'max_value': $( if [ ! -z ${VIASH_PAR_MAX_VALUE+x} ]; then echo "float(r'${VIASH_PAR_MAX_VALUE//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'zero_center': $( if [ ! -z ${VIASH_PAR_ZERO_CENTER+x} ]; then echo "r'${VIASH_PAR_ZERO_CENTER//\\'/\\'\\"\\'\\"r\\'}'.lower() == 'true'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_compression': $( if [ ! -z ${VIASH_PAR_OUTPUT_COMPRESSION+x} ]; then echo "r'${VIASH_PAR_OUTPUT_COMPRESSION//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+from compress_h5mu import write_h5ad_to_h5mu_with_compression
+
+logger = setup_logger()
+
+
+def main():
+    logger.info(
+        "Reading modality %s from .h5mu file: %s", par["modality"], par["input"]
+    )
+    data = read_h5ad(par["input"], mod=par["modality"])
+    logger.info("Scaling modality %s.", par["modality"])
+    scanpy_output = scanpy.pp.scale(
+        data,
+        layer=par["input_layer"],
+        zero_center=par["zero_center"],
+        max_value=par["max_value"],
+        copy=True,
+    )
+    output_layer_setter = (
+        partial(setattr, data, "X")
+        if not par["output_layer"]
+        else partial(setitem, data.layers, par["output_layer"])
+    )
+    output_layer_setter(
+        scanpy_output.X
+        if not par["input_layer"]
+        else scanpy_output.layers[par["input_layer"]]
+    )
+    logger.info(
+        "Writing to %s with compression %s", par["output"], par["output_compression"]
+    )
+    write_h5ad_to_h5mu_with_compression(
+        par["output"], par["input"], par["modality"], data, par["output_compression"]
+    )
+    logger.info("Finished")
+
+
+if __name__ == "__main__":
+    main()
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline/transform/scale",
+    "tag" : "v4.0.4"
+  },
+  "label" : [
+    "lowmem",
+    "lowcpu"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/nextflow.config
new file mode 100644
index 0000000..caaab9a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'transform/scale'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Scale data to unit variance and zero mean.\n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/nextflow_schema.json
new file mode 100644
index 0000000..9d20750
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/nextflow_schema.json
@@ -0,0 +1,85 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "scale",
+  "description": "Scale data to unit variance and zero mean.\n",
+  "type": "object",
+  "$defs": {
+    "arguments": {
+      "title": "Arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "List of modalities to process.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "input_layer": {
+          "type": "string",
+          "description": "Input layer with data to scale",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "output_layer": {
+          "type": "string",
+          "description": "Output layer where scaled data will be stored",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "max_value": {
+          "type": "number",
+          "description": "Clip (truncate) to this value after scaling",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "zero_center": {
+          "type": "boolean",
+          "description": "If set, omit zero-centering variables, which allows to handle sparse input efficiently.",
+          "help_text": "Type: `boolean_false`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Output h5mu file.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"output.h5mu\"`, direction: `output`. ",
+          "default": "output.h5mu"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "Compression format to use for the output AnnData and/or Mudata objects.\nBy default no compression is applied.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/transform/scale/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/.config.vsh.yaml
new file mode 100644
index 0000000..7e36003
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/.config.vsh.yaml
@@ -0,0 +1,611 @@
+name: "scanvi_scarches"
+namespace: "workflows/annotation"
+version: "v4.0.4"
+authors:
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Query Input"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the sample."
+    info: null
+    example:
+    - "foo"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    description: "Input dataset consisting of the (unlabeled) query observations.\
+      \ The dataset is expected to be pre-processed in the same way as --reference."
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality to process. Should match the modality of the --reference\
+      \ dataset."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Which layer to use for integration if .X is not to be used. Should\
+      \ match the layer of the --reference dataset."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_obs_batch_label"
+    description: "The .obs field in the input (query) dataset containing the batch\
+      \ labels."
+    info: null
+    example:
+    - "sample"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_obs_size_factor"
+    description: "Key in adata.obs for size factor information. Instead of using library\
+      \ size as a size factor,\nthe provided size factor column will be used as offset\
+      \ in the mean of the likelihood.\nAssumed to be on linear scale.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_obs_categorical_covariate"
+    description: "Keys in adata.obs that correspond to categorical data. These covariates\
+      \ can be added in\naddition to the batch covariate and are also treated as nuisance\
+      \ factors\n(i.e., the model tries to minimize their effects on the latent space).\n\
+      Thus, these should not be used for biologically-relevant factors that you do\
+      \ _not_ want to correct for.\nImportant: the order of the categorical covariates\
+      \ matters and should match the order of the covariates in the reference data.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_obs_continuous_covariate"
+    description: "Keys in adata.obs that correspond to continuous data. These covariates\
+      \ can be added in\naddition to the batch covariate and are also treated as nuisance\
+      \ factors\n(i.e., the model tries to minimize their effects on the latent space).\
+      \ Thus, these should not be\nused for biologically-relevant factors that you\
+      \ do _not_ want to correct for.\nImportant: the order of the continuous covariates\
+      \ matters and should match the order of the covariates in the reference data.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_var_gene_names"
+    description: ".var column containing gene names. By default, use the index."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--sanitize_ensembl_ids"
+    description: "Whether to sanitize ensembl ids by removing version numbers."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Reference input"
+  arguments:
+  - type: "file"
+    name: "--reference"
+    description: "Reference dataset consisting of the labeled observations to train\
+      \ the KNN classifier on. The dataset is expected to be pre-processed in the\
+      \ same way as the --input query dataset."
+    info: null
+    example:
+    - "reference.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_obs_target"
+    description: "The `.obs` key containing the target labels."
+    info: null
+    example:
+    - "cell_type"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_obs_batch_label"
+    description: "The .obs field in the reference dataset containing the batch labels."
+    info: null
+    example:
+    - "sample"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_obs_size_factor"
+    description: "Key in adata.obs for size factor information. Instead of using library\
+      \ size as a size factor,\nthe provided size factor column will be used as offset\
+      \ in the mean of the likelihood.\nAssumed to be on linear scale.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_obs_categorical_covariate"
+    description: "Keys in adata.obs that correspond to categorical data. These covariates\
+      \ can be added in\naddition to the batch covariate and are also treated as nuisance\
+      \ factors\n(i.e., the model tries to minimize their effects on the latent space).\n\
+      Thus, these should not be used for biologically-relevant factors that you do\
+      \ _not_ want to correct for.\nImportant: the order of the categorical covariates\
+      \ matters and should match the order of the covariates in the query data.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_obs_continuous_covariate"
+    description: "Keys in adata.obs that correspond to continuous data. These covariates\
+      \ can be added in\naddition to the batch covariate and are also treated as nuisance\
+      \ factors\n(i.e., the model tries to minimize their effects on the latent space).\
+      \ Thus, these should not be\nused for biologically-relevant factors that you\
+      \ do _not_ want to correct for.\nImportant: the order of the continuous covariates\
+      \ matters and should match the order of the covariates in the query data.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--unlabeled_category"
+    description: "Value in the --reference_obs_batch_label field that indicates unlabeled\
+      \ observations\n"
+    info: null
+    default:
+    - "Unknown"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_var_hvg"
+    description: ".var column containing highly variable genes. If not provided, genes\
+      \ will not be subset."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_var_gene_names"
+    description: ".var column containing gene names. By default, use the index."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "scVI, scANVI and scArches training options"
+  arguments:
+  - type: "boolean"
+    name: "--early_stopping"
+    description: "Whether to perform early stopping with respect to the validation\
+      \ set."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--early_stopping_monitor"
+    description: "Metric logged during validation set epoch."
+    info: null
+    default:
+    - "elbo_validation"
+    required: false
+    choices:
+    - "elbo_validation"
+    - "reconstruction_loss_validation"
+    - "kl_local_validation"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--early_stopping_patience"
+    description: "Number of validation epochs with no improvement after which training\
+      \ will be stopped."
+    info: null
+    default:
+    - 45
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--early_stopping_min_delta"
+    description: "Minimum change in the monitored quantity to qualify as an improvement,\
+      \ i.e. an absolute change of less than min_delta, will count as no improvement."
+    info: null
+    default:
+    - 0.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_epochs"
+    description: "Number of passes through the dataset, defaults to (20000 / number\
+      \ of cells) * 400 or 400; whichever is smallest."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--reduce_lr_on_plateau"
+    description: "Whether to monitor validation loss and reduce learning rate when\
+      \ validation set `lr_scheduler_metric` plateaus."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--lr_factor"
+    description: "Factor to reduce learning rate."
+    info: null
+    default:
+    - 0.6
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--lr_patience"
+    description: "Number of epochs with no improvement after which learning rate will\
+      \ be reduced."
+    info: null
+    default:
+    - 30.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Leiden clustering options"
+  arguments:
+  - type: "double"
+    name: "--leiden_resolution"
+    description: "Control the coarseness of the clustering. Higher values lead to\
+      \ more clusters."
+    info: null
+    default:
+    - 1.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Neighbor classifier arguments"
+  arguments:
+  - type: "string"
+    name: "--knn_weights"
+    description: "Weight function used in prediction. Possible values are:\n`uniform`\
+      \ (all points in each neighborhood are weighted equally) or \n`distance` (weight\
+      \ points by the inverse of their distance)\n"
+    info: null
+    default:
+    - "uniform"
+    required: false
+    choices:
+    - "uniform"
+    - "distance"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--knn_n_neighbors"
+    description: "The number of neighbors to use in k-neighbor graph structure used\
+      \ for fast approximate nearest neighbor search with PyNNDescent. \nLarger values\
+      \ will result in more accurate search results at the cost of computation time.\n"
+    info: null
+    default:
+    - 15
+    required: false
+    min: 5
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "The query data in .h5mu format with predicted labels predicted from\
+      \ the classifier trained on the reference."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_predictions"
+    description: "In which .obs slot to store the predicted labels."
+    info: null
+    default:
+    - "scanvi_pred"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_probability"
+    description: "In which. obs slot to store the probabilities of the predicted labels."
+    info: null
+    default:
+    - "scanvi_probabilities"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obsm_integrated"
+    description: "In which .obsm slot to store the integrated embedding."
+    info: null
+    default:
+    - "X_integrated_scanvi"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_compression"
+    description: "The compression format to be used on the output h5mu object.\n"
+    info: null
+    example:
+    - "gzip"
+    required: false
+    choices:
+    - "gzip"
+    - "lzf"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_model"
+    description: "Path to the resulting scANVI model that was updated with query data."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Cell type annotation workflow using ScanVI with scArches for reference\
+  \ mapping."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "TS_Blood_filtered.h5mu"
+- type: "file"
+  path: "pbmc_1k_protein_v3_mms.h5mu"
+info:
+  name: "scANVI - scArches workflow"
+  test_dependencies:
+  - name: "scanvi_scarches_test"
+    namespace: "test_workflows/annotation"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "integrate/scvi"
+  repository:
+    type: "local"
+- name: "annotate/scanvi"
+  repository:
+    type: "local"
+- name: "integrate/scarches"
+  repository:
+    type: "local"
+- name: "workflows/multiomics/neighbors_leiden_umap"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/annotation/scanvi_scarches/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/annotation/scanvi_scarches"
+  executable: "target/nextflow/workflows/annotation/scanvi_scarches/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/integrate/scvi"
+  - "target/nextflow/annotate/scanvi"
+  - "target/nextflow/integrate/scarches"
+  - "target/nextflow/workflows/multiomics/neighbors_leiden_umap"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/main.nf
new file mode 100644
index 0000000..bac9e4b
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/main.nf
@@ -0,0 +1,4011 @@
+// scanvi_scarches v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dorien Roosen (author, maintainer)
+//  * Weiwei Schultz (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "scanvi_scarches",
+  "namespace" : "workflows/annotation",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Query Input",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the sample.",
+          "example" : [
+            "foo"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Input dataset consisting of the (unlabeled) query observations. The dataset is expected to be pre-processed in the same way as --reference.",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality to process. Should match the modality of the --reference dataset.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Which layer to use for integration if .X is not to be used. Should match the layer of the --reference dataset.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_obs_batch_label",
+          "description" : "The .obs field in the input (query) dataset containing the batch labels.",
+          "example" : [
+            "sample"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_obs_size_factor",
+          "description" : "Key in adata.obs for size factor information. Instead of using library size as a size factor,\nthe provided size factor column will be used as offset in the mean of the likelihood.\nAssumed to be on linear scale.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_obs_categorical_covariate",
+          "description" : "Keys in adata.obs that correspond to categorical data. These covariates can be added in\naddition to the batch covariate and are also treated as nuisance factors\n(i.e., the model tries to minimize their effects on the latent space).\nThus, these should not be used for biologically-relevant factors that you do _not_ want to correct for.\nImportant: the order of the categorical covariates matters and should match the order of the covariates in the reference data.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_obs_continuous_covariate",
+          "description" : "Keys in adata.obs that correspond to continuous data. These covariates can be added in\naddition to the batch covariate and are also treated as nuisance factors\n(i.e., the model tries to minimize their effects on the latent space). Thus, these should not be\nused for biologically-relevant factors that you do _not_ want to correct for.\nImportant: the order of the continuous covariates matters and should match the order of the covariates in the reference data.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_var_gene_names",
+          "description" : ".var column containing gene names. By default, use the index.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--sanitize_ensembl_ids",
+          "description" : "Whether to sanitize ensembl ids by removing version numbers.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Reference input",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--reference",
+          "description" : "Reference dataset consisting of the labeled observations to train the KNN classifier on. The dataset is expected to be pre-processed in the same way as the --input query dataset.",
+          "example" : [
+            "reference.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_obs_target",
+          "description" : "The `.obs` key containing the target labels.",
+          "example" : [
+            "cell_type"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_obs_batch_label",
+          "description" : "The .obs field in the reference dataset containing the batch labels.",
+          "example" : [
+            "sample"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_obs_size_factor",
+          "description" : "Key in adata.obs for size factor information. Instead of using library size as a size factor,\nthe provided size factor column will be used as offset in the mean of the likelihood.\nAssumed to be on linear scale.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_obs_categorical_covariate",
+          "description" : "Keys in adata.obs that correspond to categorical data. These covariates can be added in\naddition to the batch covariate and are also treated as nuisance factors\n(i.e., the model tries to minimize their effects on the latent space).\nThus, these should not be used for biologically-relevant factors that you do _not_ want to correct for.\nImportant: the order of the categorical covariates matters and should match the order of the covariates in the query data.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_obs_continuous_covariate",
+          "description" : "Keys in adata.obs that correspond to continuous data. These covariates can be added in\naddition to the batch covariate and are also treated as nuisance factors\n(i.e., the model tries to minimize their effects on the latent space). Thus, these should not be\nused for biologically-relevant factors that you do _not_ want to correct for.\nImportant: the order of the continuous covariates matters and should match the order of the covariates in the query data.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--unlabeled_category",
+          "description" : "Value in the --reference_obs_batch_label field that indicates unlabeled observations\n",
+          "default" : [
+            "Unknown"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_var_hvg",
+          "description" : ".var column containing highly variable genes. If not provided, genes will not be subset.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_var_gene_names",
+          "description" : ".var column containing gene names. By default, use the index.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "scVI, scANVI and scArches training options",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--early_stopping",
+          "description" : "Whether to perform early stopping with respect to the validation set.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--early_stopping_monitor",
+          "description" : "Metric logged during validation set epoch.",
+          "default" : [
+            "elbo_validation"
+          ],
+          "required" : false,
+          "choices" : [
+            "elbo_validation",
+            "reconstruction_loss_validation",
+            "kl_local_validation"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--early_stopping_patience",
+          "description" : "Number of validation epochs with no improvement after which training will be stopped.",
+          "default" : [
+            45
+          ],
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--early_stopping_min_delta",
+          "description" : "Minimum change in the monitored quantity to qualify as an improvement, i.e. an absolute change of less than min_delta, will count as no improvement.",
+          "default" : [
+            0.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_epochs",
+          "description" : "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--reduce_lr_on_plateau",
+          "description" : "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--lr_factor",
+          "description" : "Factor to reduce learning rate.",
+          "default" : [
+            0.6
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--lr_patience",
+          "description" : "Number of epochs with no improvement after which learning rate will be reduced.",
+          "default" : [
+            30.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Leiden clustering options",
+      "arguments" : [
+        {
+          "type" : "double",
+          "name" : "--leiden_resolution",
+          "description" : "Control the coarseness of the clustering. Higher values lead to more clusters.",
+          "default" : [
+            1.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Neighbor classifier arguments",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--knn_weights",
+          "description" : "Weight function used in prediction. Possible values are:\n`uniform` (all points in each neighborhood are weighted equally) or \n`distance` (weight points by the inverse of their distance)\n",
+          "default" : [
+            "uniform"
+          ],
+          "required" : false,
+          "choices" : [
+            "uniform",
+            "distance"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--knn_n_neighbors",
+          "description" : "The number of neighbors to use in k-neighbor graph structure used for fast approximate nearest neighbor search with PyNNDescent. \nLarger values will result in more accurate search results at the cost of computation time.\n",
+          "default" : [
+            15
+          ],
+          "required" : false,
+          "min" : 5,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "The query data in .h5mu format with predicted labels predicted from the classifier trained on the reference.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_predictions",
+          "description" : "In which .obs slot to store the predicted labels.",
+          "default" : [
+            "scanvi_pred"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_probability",
+          "description" : "In which. obs slot to store the probabilities of the predicted labels.",
+          "default" : [
+            "scanvi_probabilities"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obsm_integrated",
+          "description" : "In which .obsm slot to store the integrated embedding.",
+          "default" : [
+            "X_integrated_scanvi"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_compression",
+          "description" : "The compression format to be used on the output h5mu object.\n",
+          "example" : [
+            "gzip"
+          ],
+          "required" : false,
+          "choices" : [
+            "gzip",
+            "lzf"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_model",
+          "description" : "Path to the resulting scANVI model that was updated with query data.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Cell type annotation workflow using ScanVI with scArches for reference mapping.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/annotation_test_data/TS_Blood_filtered.h5mu"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_mms.h5mu"
+    }
+  ],
+  "info" : {
+    "name" : "scANVI - scArches workflow",
+    "test_dependencies" : [
+      {
+        "name" : "scanvi_scarches_test",
+        "namespace" : "test_workflows/annotation"
+      }
+    ]
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "integrate/scvi",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "annotate/scanvi",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "integrate/scarches",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/multiomics/neighbors_leiden_umap",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/annotation/scanvi_scarches/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/annotation/scanvi_scarches",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { scvi } from "${meta.resources_dir}/../../../../nextflow/integrate/scvi/main.nf"
+include { scanvi } from "${meta.resources_dir}/../../../../nextflow/annotate/scanvi/main.nf"
+include { scarches } from "${meta.resources_dir}/../../../../nextflow/integrate/scarches/main.nf"
+include { neighbors_leiden_umap } from "${meta.resources_dir}/../../../../nextflow/workflows/multiomics/neighbors_leiden_umap/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+    input_ch
+
+  main:
+
+    output_ch = input_ch
+        // Set aside the output for this workflow to avoid conflicts
+        | map {id, state -> 
+        def new_state = state + ["workflow_output": state.output, "workflow_output_model": state.output_model]
+        [id, new_state]
+        }
+
+        // Integrate & generate scvi model from the reference data
+        | scvi.run(
+          fromState: [
+              "input": "reference",
+              "modality": "modality",
+              "input_layer": "layer",
+              "obs_batch": "reference_obs_batch_label",
+              "var_input": "reference_var_hvg",
+              "var_gene_names": "reference_var_gene_names",
+              "obs_size_factor": "reference_obs_size_factor",
+              "obs_categorical_covariate": "reference_obs_categorical_covariate",
+              "obs_continuous_covariate": "reference_obs_continuous_covariate",
+              "early_stopping": "early_stopping",
+              "early_stopping_monitor": "early_stopping_monitor",
+              "early_stopping_patience": "early_stopping_patience",
+              "early_stopping_min_delta": "early_stopping_min_delta",
+              "max_epochs": "max_epochs",
+              "reduce_lr_on_plateau": "reduce_lr_on_plateau",
+              "lr_factor": "lr_factor",
+              "lr_patience": "lr_patience",
+              "sanitize_ensembl_ids": "sanitize_ensembl_ids"
+          ],
+          args: [
+              "obsm_output": "X_integrated_scvi"
+          ],
+          toState: [
+              "reference": "output",
+              "output_scvi_model": "output_model"
+          ]
+        )
+        
+        // Create scanvi model from the scvi reference model and integrate reference data
+        | scanvi.run(
+          fromState: [
+              "input": "reference",
+              "modality": "modality",
+              "input_layer": "layer",
+              "var_input": "reference_var_hvg",
+              "var_gene_names": "reference_var_gene_names",
+              "obs_labels": "reference_obs_target",
+              "unlabeled_category": "unlabeled_category",
+              "scvi_model": "output_scvi_model",
+              "obsm_output": "output_obsm_integrated",
+              "obs_output_predictions": "output_obs_predictions",
+              "obs_output_probabilities": "output_obs_probability",
+              "early_stopping": "early_stopping",
+              "early_stopping_monitor": "early_stopping_monitor",
+              "early_stopping_patience": "early_stopping_patience",
+              "early_stopping_min_delta": "early_stopping_min_delta",
+              "max_epochs": "max_epochs",
+              "reduce_lr_on_plateau": "reduce_lr_on_plateau",
+              "lr_factor": "lr_factor",
+              "lr_patience": "lr_patience",
+              "sanitize_ensembl_ids": "sanitize_ensembl_ids"
+          ],
+          toState: [
+              "reference": "output",
+              "scanvi_model": "output_model"
+          ]
+        )
+
+        // Update scANVI reference model with query data and integrate+annotate query data
+        | scarches.run(
+          fromState: [
+              "input": "input",
+              "layer": "layer",
+              "modality": "modality",
+              "input_obs_batch": "input_obs_batch_label",
+              "input_var_gene_names": "input_var_gene_names",
+              "input_obs_size_factor": "input_obs_size_factor",
+              "input_obs_categorical_covariate": "input_obs_categorical_covariate",
+              "input_obs_continuous_covariate": "input_obs_continuous_covariate",
+              "reference": "scanvi_model",
+              "obsm_output": "output_obsm_integrated",
+              "obs_output_predictions": "output_obs_predictions",
+              "obs_output_probabilities": "output_obs_probability",
+              "early_stopping": "early_stopping",
+              "early_stopping_monitor": "early_stopping_monitor",
+              "early_stopping_patience": "early_stopping_patience",
+              "early_stopping_min_delta": "early_stopping_min_delta",
+              "max_epochs": "max_epochs",
+              "reduce_lr_on_plateau": "reduce_lr_on_plateau",
+              "lr_factor": "lr_factor",
+              "lr_patience": "lr_patience",
+              "output": "workflow_output",
+              "model_output": "workflow_output_model",
+              "sanitize_ensembl_ids": "sanitize_ensembl_ids"
+          ],
+          toState: [
+              "input": "output",
+              "output_model": "model_output"
+          ]
+        )
+        | view {"After scarches: $it"}
+        // Perform neighbors calculation, leiden clustering and umap dimred on the integrated query data
+        | neighbors_leiden_umap.run(
+          fromState: [
+              "input": "input",
+              "modality": "modality",
+              "obsm_input": "output_obsm_integrated",
+              "output": "workflow_output",
+              "leiden_resolution": "leiden_resolution"
+          ],
+          args: [
+              "uns_neighbors": "scanvi_integration_neighbors",
+              "obsp_neighbor_distances": "scanvi_integration_distances",
+              "obsp_neighbor_connectivities": "scanvi_integration_connectivities",
+              "obs_cluster": "scanvi_integration_leiden",
+              "obsm_umap": "X_scanvi_umap"
+          ],
+          toState: [ "output": "output" ]
+        )
+        | setState(["output", "output_model"])
+
+  emit:
+    output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/nextflow.config
new file mode 100644
index 0000000..e38033e
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/annotation/scanvi_scarches'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Cell type annotation workflow using ScanVI with scArches for reference mapping.'
+  author = 'Dorien Roosen, Weiwei Schultz'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/nextflow_schema.json
new file mode 100644
index 0000000..df1e269
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/nextflow_schema.json
@@ -0,0 +1,318 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "scanvi_scarches",
+  "description": "Cell type annotation workflow using ScanVI with scArches for reference mapping.",
+  "type": "object",
+  "$defs": {
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "The query data in .h5mu format with predicted labels predicted from the classifier trained on the reference.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "output_obs_predictions": {
+          "type": "string",
+          "description": "In which .obs slot to store the predicted labels.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scanvi_pred\"`. ",
+          "default": "scanvi_pred"
+        },
+        "output_obs_probability": {
+          "type": "string",
+          "description": "In which",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scanvi_probabilities\"`. ",
+          "default": "scanvi_probabilities"
+        },
+        "output_obsm_integrated": {
+          "type": "string",
+          "description": "In which .obsm slot to store the integrated embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_integrated_scanvi\"`. ",
+          "default": "X_integrated_scanvi"
+        },
+        "output_compression": {
+          "type": "string",
+          "description": "The compression format to be used on the output h5mu object.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gzip\"`, choices: ``gzip`, `lzf``. ",
+          "enum": [
+            "gzip",
+            "lzf"
+          ]
+        },
+        "output_model": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to the resulting scANVI model that was updated with query data.",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output_model\"`, direction: `output`. ",
+          "default": "$id.$key.output_model"
+        }
+      }
+    },
+    "query input": {
+      "title": "Query Input",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "ID of the sample.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"foo\"`. "
+        },
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input dataset consisting of the (unlabeled) query observations",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality to process",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "layer": {
+          "type": "string",
+          "description": "Which layer to use for integration if .X is not to be used",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "input_obs_batch_label": {
+          "type": "string",
+          "description": "The .obs field in the input (query) dataset containing the batch labels.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"sample\"`. "
+        },
+        "input_obs_size_factor": {
+          "type": "string",
+          "description": "Key in adata.obs for size factor information",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "input_obs_categorical_covariate": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys in adata.obs that correspond to categorical data",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "input_obs_continuous_covariate": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys in adata.obs that correspond to continuous data",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "input_var_gene_names": {
+          "type": "string",
+          "description": ".var column containing gene names",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "sanitize_ensembl_ids": {
+          "type": "boolean",
+          "description": "Whether to sanitize ensembl ids by removing version numbers.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "reference input": {
+      "title": "Reference input",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "reference": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Reference dataset consisting of the labeled observations to train the KNN classifier on",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"reference.h5mu\"`. "
+        },
+        "reference_obs_target": {
+          "type": "string",
+          "description": "The `.obs` key containing the target labels.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"cell_type\"`. "
+        },
+        "reference_obs_batch_label": {
+          "type": "string",
+          "description": "The .obs field in the reference dataset containing the batch labels.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"sample\"`. "
+        },
+        "reference_obs_size_factor": {
+          "type": "string",
+          "description": "Key in adata.obs for size factor information",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "reference_obs_categorical_covariate": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys in adata.obs that correspond to categorical data",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "reference_obs_continuous_covariate": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys in adata.obs that correspond to continuous data",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "unlabeled_category": {
+          "type": "string",
+          "description": "Value in the --reference_obs_batch_label field that indicates unlabeled observations\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"Unknown\"`. ",
+          "default": "Unknown"
+        },
+        "reference_var_hvg": {
+          "type": "string",
+          "description": ".var column containing highly variable genes",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "reference_var_gene_names": {
+          "type": "string",
+          "description": ".var column containing gene names",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "scvi, scanvi and scarches training options": {
+      "title": "scVI, scANVI and scArches training options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "early_stopping": {
+          "type": "boolean",
+          "description": "Whether to perform early stopping with respect to the validation set.",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        },
+        "early_stopping_monitor": {
+          "type": "string",
+          "description": "Metric logged during validation set epoch.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"elbo_validation\"`, choices: ``elbo_validation`, `reconstruction_loss_validation`, `kl_local_validation``. ",
+          "enum": [
+            "elbo_validation",
+            "reconstruction_loss_validation",
+            "kl_local_validation"
+          ],
+          "default": "elbo_validation"
+        },
+        "early_stopping_patience": {
+          "type": "integer",
+          "description": "Number of validation epochs with no improvement after which training will be stopped.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `45`. ",
+          "default": 45
+        },
+        "early_stopping_min_delta": {
+          "type": "number",
+          "description": "Minimum change in the monitored quantity to qualify as an improvement, i.e",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.0`. ",
+          "default": 0.0
+        },
+        "max_epochs": {
+          "type": "integer",
+          "description": "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest.",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "reduce_lr_on_plateau": {
+          "type": "boolean",
+          "description": "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "lr_factor": {
+          "type": "number",
+          "description": "Factor to reduce learning rate.",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.6`. ",
+          "default": 0.6
+        },
+        "lr_patience": {
+          "type": "number",
+          "description": "Number of epochs with no improvement after which learning rate will be reduced.",
+          "help_text": "Type: `double`, multiple: `False`, default: `30.0`. ",
+          "default": 30.0
+        }
+      }
+    },
+    "leiden clustering options": {
+      "title": "Leiden clustering options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "leiden_resolution": {
+          "type": "array",
+          "items": {
+            "type": "number"
+          },
+          "description": "Control the coarseness of the clustering",
+          "help_text": "Type: `double`, multiple: `True`, default: `[1.0]`. ",
+          "default": [
+            1.0
+          ]
+        }
+      }
+    },
+    "neighbor classifier arguments": {
+      "title": "Neighbor classifier arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "knn_weights": {
+          "type": "string",
+          "description": "Weight function used in prediction",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"uniform\"`, choices: ``uniform`, `distance``. ",
+          "enum": [
+            "uniform",
+            "distance"
+          ],
+          "default": "uniform"
+        },
+        "knn_n_neighbors": {
+          "type": "integer",
+          "description": "The number of neighbors to use in k-neighbor graph structure used for fast approximate nearest neighbor search with PyNNDescent",
+          "help_text": "Type: `integer`, multiple: `False`, default: `15`. ",
+          "default": 15
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/query input"
+    },
+    {
+      "$ref": "#/$defs/reference input"
+    },
+    {
+      "$ref": "#/$defs/scvi, scanvi and scarches training options"
+    },
+    {
+      "$ref": "#/$defs/leiden clustering options"
+    },
+    {
+      "$ref": "#/$defs/neighbor classifier arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/.config.vsh.yaml
new file mode 100644
index 0000000..3e3f0c8
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/.config.vsh.yaml
@@ -0,0 +1,276 @@
+name: "gdo_singlesample"
+namespace: "workflows/gdo"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Input"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the sample."
+    info: null
+    example:
+    - "foo"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    description: "Path to the sample."
+    info: null
+    example:
+    - "dataset.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Input layer to start from. By default, .X will be used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Output"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Filtering options"
+  arguments:
+  - type: "integer"
+    name: "--min_counts"
+    description: "Minimum number of counts captured per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_counts"
+    description: "Maximum number of counts captured per cell."
+    info: null
+    example:
+    - 5000000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_guides_per_cell"
+    description: "Minimum of non-zero values per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_guides_per_cell"
+    description: "Maximum of non-zero values per cell."
+    info: null
+    example:
+    - 1500000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_cells_per_guide"
+    description: "Minimum of non-zero values per gene."
+    info: null
+    example:
+    - 3
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Processing unimodal single-sample guide-derived oligonucleotide (GDO)\
+  \ data."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "10x_5k_lung_crispr"
+info:
+  name: "GDO Singlesample"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "filter/filter_with_counts"
+  repository:
+    type: "local"
+- name: "filter/do_filter"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/gdo/gdo_singlesample/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/gdo/gdo_singlesample"
+  executable: "target/nextflow/workflows/gdo/gdo_singlesample/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/filter/filter_with_counts"
+  - "target/nextflow/filter/do_filter"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/main.nf
new file mode 100644
index 0000000..0464bf9
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/main.nf
@@ -0,0 +1,3570 @@
+// gdo_singlesample v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "gdo_singlesample",
+  "namespace" : "workflows/gdo",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Input",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the sample.",
+          "example" : [
+            "foo"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Path to the sample.",
+          "example" : [
+            "dataset.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Input layer to start from. By default, .X will be used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Output",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Filtering options",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--min_counts",
+          "description" : "Minimum number of counts captured per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_counts",
+          "description" : "Maximum number of counts captured per cell.",
+          "example" : [
+            5000000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_guides_per_cell",
+          "description" : "Minimum of non-zero values per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_guides_per_cell",
+          "description" : "Maximum of non-zero values per cell.",
+          "example" : [
+            1500000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_cells_per_guide",
+          "description" : "Minimum of non-zero values per gene.",
+          "example" : [
+            3
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Processing unimodal single-sample guide-derived oligonucleotide (GDO) data.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_lung_crispr"
+    }
+  ],
+  "info" : {
+    "name" : "GDO Singlesample"
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "filter/filter_with_counts",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "filter/do_filter",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/gdo/gdo_singlesample/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/gdo/gdo_singlesample",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { filter_with_counts } from "${meta.resources_dir}/../../../../nextflow/filter/filter_with_counts/main.nf"
+include { do_filter } from "${meta.resources_dir}/../../../../nextflow/filter/do_filter/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+  output_ch = input_ch
+    // Set aside the output for this workflow to avoid conflicts
+    | map {id, state -> 
+      def new_state = state + ["workflow_output": state.output]
+      [id, new_state]
+    }
+    // cell filtering
+    | filter_with_counts.run(
+      key: "gdo_filter_with_counts",
+      fromState: { id, state ->
+        [
+          "input": state.input,
+          "modality": "gdo",
+          "obs_name_filter": "filter_with_counts",
+          "var_name_filter": "filter_with_counts",
+          "min_counts": state.min_counts,
+          "max_counts": state.max_counts,
+          "min_genes_per_cell": state.min_guides_per_cell,
+          "max_genes_per_cell": state.max_guides_per_cell,
+          "min_cells_per_gene": state.min_cells_per_guide,
+          "layer": state.layer,
+        ]
+      },
+      toState: ["input": "output"]
+    )
+    | do_filter.run(
+      key: "gdo_do_filter",
+      fromState : { id, state ->
+        // do_filter does not need a layer argument because it filters all layers
+        // from a modality.
+        def newState = [
+          "input": state.input,
+          "obs_filter": "filter_with_counts",
+          "modality": "gdo",
+          "var_filter": "filter_with_counts",
+          "output_compression": "gzip",
+          "output": state.workflow_output
+        ]
+        return newState
+      },
+      toState: ["output": "output"]
+    )
+    | setState(["output"])
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/nextflow.config
new file mode 100644
index 0000000..faff528
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/gdo/gdo_singlesample'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Processing unimodal single-sample guide-derived oligonucleotide (GDO) data.'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/nextflow_schema.json
new file mode 100644
index 0000000..1a0ffda
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/nextflow_schema.json
@@ -0,0 +1,104 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "gdo_singlesample",
+  "description": "Processing unimodal single-sample guide-derived oligonucleotide (GDO) data.",
+  "type": "object",
+  "$defs": {
+    "input": {
+      "title": "Input",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "ID of the sample.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"foo\"`. "
+        },
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the sample.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"dataset.h5mu\"`. "
+        },
+        "layer": {
+          "type": "string",
+          "description": "Input layer to start from",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "output": {
+      "title": "Output",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "filtering options": {
+      "title": "Filtering options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "min_counts": {
+          "type": "integer",
+          "description": "Minimum number of counts captured per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "max_counts": {
+          "type": "integer",
+          "description": "Maximum number of counts captured per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `5000000`. "
+        },
+        "min_guides_per_cell": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "max_guides_per_cell": {
+          "type": "integer",
+          "description": "Maximum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `1500000`. "
+        },
+        "min_cells_per_guide": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per gene.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3`. "
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/input"
+    },
+    {
+      "$ref": "#/$defs/output"
+    },
+    {
+      "$ref": "#/$defs/filtering options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/gdo/gdo_singlesample/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/.config.vsh.yaml
new file mode 100644
index 0000000..568d88b
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/.config.vsh.yaml
@@ -0,0 +1,958 @@
+name: "cellranger_multi"
+namespace: "workflows/ingestion"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Input files"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "The FASTQ files to be analyzed. FASTQ files should conform to the\
+      \ naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane\
+      \ Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Feature type-specific input files"
+  description: "Helper functionality to allow feature type-specific input files, without\
+    \ the need to specify\nlibrary_type or library_id. The library_id will be inferred\
+    \ from the input paths.\n"
+  arguments:
+  - type: "file"
+    name: "--gex_input"
+    description: "The FASTQ files to be analyzed for Gene Expression. FASTQ files\
+      \ should conform to the\nnaming conventions of bcl2fastq and mkfastq:\n`[Sample\
+      \ Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--abc_input"
+    description: "The FASTQ files to be analyzed for Antibody Capture. FASTQ files\
+      \ should conform to \nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample\
+      \ Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--cgc_input"
+    description: "The FASTQ files to be analyzed for CRISPR Guide Capture. FASTQ files\
+      \ should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample\
+      \ Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--mux_input"
+    description: "The FASTQ files to be analyzed for Multiplexing Capture. FASTQ files\
+      \ should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample\
+      \ Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--vdj_input"
+    description: "The FASTQ files to be analyzed for VDJ. FASTQ files should conform\
+      \ to the\nnaming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample\
+      \ Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--vdj_t_input"
+    description: "The FASTQ files to be analyzed for VDJ-T. FASTQ files should conform\
+      \ to the naming\nconventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample\
+      \ Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--vdj_t_gd_input"
+    description: "The FASTQ files to be analyzed for VDJ-T-GD. FASTQ files should\
+      \ conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample\
+      \ Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--vdj_b_input"
+    description: "The FASTQ files to be analyzed for VDJ-B. FASTQ files should conform\
+      \ to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample\
+      \ Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--agc_input"
+    description: "The FASTQ files to be analyzed for Antigen Capture. FASTQ files\
+      \ should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample\
+      \ Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Library arguments"
+  arguments:
+  - type: "string"
+    name: "--library_id"
+    description: "The Illumina sample name to analyze. This must exactly match the\
+      \ 'Sample Name'part\nof the FASTQ files specified in the `--input` argument.\n"
+    info: null
+    example:
+    - "mysample1"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--library_type"
+    description: "The underlying feature type of the library.\n"
+    info: null
+    example:
+    - "Gene Expression"
+    required: false
+    choices:
+    - "Gene Expression"
+    - "VDJ"
+    - "VDJ-T"
+    - "VDJ-B"
+    - "VDJ-T-GD"
+    - "Antibody Capture"
+    - "CRISPR Guide Capture"
+    - "Multiplexing Capture"
+    - "Antigen Capture"
+    - "Custom"
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--library_subsample"
+    description: "The rate at which reads from the provided FASTQ files are sampled.\n\
+      Must be strictly greater than 0 and less than or equal to 1.\n"
+    info: null
+    example:
+    - "0.5"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--library_lanes"
+    description: "Lanes associated with this sample. Defaults to using all lanes."
+    info: null
+    example:
+    - "1-4"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--library_chemistry"
+    description: "Only applicable to FRP. Library-specific assay configuration. By\
+      \ default,\nthe assay configuration is detected automatically. Typically, users\
+      \ will\nnot need to specify a chemistry.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Sample parameters"
+  arguments:
+  - type: "string"
+    name: "--sample_ids"
+    alternatives:
+    - "--cell_multiplex_sample_id"
+    description: "A name to identify a multiplexed sample. Must be alphanumeric with\
+      \ hyphens and/or underscores,\nand less than 64 characters. Required for Cell\
+      \ Multiplexing libraries.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--sample_description"
+    alternatives:
+    - "--cell_multiplex_description"
+    description: "A description for the sample."
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--sample_expect_cells"
+    description: "Expected number of recovered cells, used as input to cell calling\
+      \ algorithm.\n"
+    info: null
+    example:
+    - 3000
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--sample_force_cells"
+    description: "Force pipeline to use this number of cells, bypassing cell detection.\n"
+    info: null
+    example:
+    - 3000
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Feature Barcode library specific arguments"
+  arguments:
+  - type: "file"
+    name: "--feature_reference"
+    description: "Path to the Feature reference CSV file, declaring Feature Barcode\
+      \ constructs and associated barcodes.\nRequired only for Antibody Capture or\
+      \ CRISPR Guide Capture libraries.\nSee https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/using/feature-bc-analysis#feature-ref\
+      \ for more information.\"\n"
+    info: null
+    example:
+    - "feature_reference.csv"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--feature_r1_length"
+    description: "Limit the length of the input Read 1 sequence of V(D)J libraries\
+      \ to the first N bases,\nwhere N is the user-supplied value. Note that the length\
+      \ includes the Barcode and UMI\nsequences so do not set this below 26.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--feature_r2_length"
+    description: "Limit the length of the input Read 2 sequence of V(D)J libraries\
+      \ to the first N bases,\nwhere N is a user-supplied value. Trimming occurs before\
+      \ sequencing metrics are computed\nand therefore, limiting the length of Read\
+      \ 2 may affect Q30 scores.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_crispr_umi"
+    description: "Set the minimum number of CRISPR guide RNA UMIs required for protospacer\
+      \ detection.\nIf a lower or higher sensitivity is desired for detection, this\
+      \ value can be customized\naccording to specific experimental needs. Applicable\
+      \ only to datasets that include a\nCRISPR Guide Capture library.\n"
+    info: null
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Gene expression arguments"
+  description: "Arguments relevant to the analysis of gene expression data."
+  arguments:
+  - type: "file"
+    name: "--gex_reference"
+    description: "Genome refence index built by Cell Ranger mkref."
+    info: null
+    example:
+    - "reference_genome.tar.gz"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--gex_secondary_analysis"
+    description: "Whether or not to run the secondary analysis e.g. clustering."
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--gex_generate_bam"
+    description: "Whether to generate a BAM file."
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--tenx_cloud_token_path"
+    description: "The 10x Cloud Analysis user token used to enable cell annotation."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--cell_annotation_model"
+    description: "\"Cell annotation model to use. If auto, uses the default model\
+      \ for the species.\nIf not given, does not run cell annotation.\"\n"
+    info: null
+    required: false
+    choices:
+    - "auto"
+    - "human_pca_v1_beta"
+    - "mouse_pca_v1_beta"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gex_expect_cells"
+    description: "Expected number of recovered cells, used as input to cell calling\
+      \ algorithm.\n"
+    info: null
+    example:
+    - 3000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gex_force_cells"
+    description: "Force pipeline to use this number of cells, bypassing cell detection.\n"
+    info: null
+    example:
+    - 3000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--gex_include_introns"
+    description: "Whether or not to include intronic reads in counts.\nThis option\
+      \ does not apply to Fixed RNA Profiling analysis.\n"
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gex_r1_length"
+    description: "Limit the length of the input Read 1 sequence of V(D)J libraries\
+      \ to the first N bases,\nwhere N is the user-supplied value. Note that the length\
+      \ includes the Barcode and UMI\nsequences so do not set this below 26.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gex_r2_length"
+    description: "Limit the length of the input Read 2 sequence of V(D)J libraries\
+      \ to the first N bases,\nwhere N is a user-supplied value. Trimming occurs before\
+      \ sequencing metrics are computed\nand therefore, limiting the length of Read\
+      \ 2 may affect Q30 scores.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--gex_chemistry"
+    description: "Assay configuration. Either specify a single value which will be\
+      \ applied to all libraries,\nor a number of values that is equal to the number\
+      \ of libararies. The latter is only applicable\nto only applicable to Fixed\
+      \ RNA Profiling.\n  - auto: Chemistry autodetection (default)\n  - threeprime:\
+      \ Single Cell 3'\n  - SC3Pv1, SC3Pv2, SC3Pv3(-polyA), SC3Pv4(-polyA): Single\
+      \ Cell 3' v1, v2, v3, or v4\n  - SC3Pv3HT(-polyA): Single Cell 3' v3.1 HT\n\
+      \  - SC-FB: Single Cell Antibody-only 3' v2 or 5'\n  - fiveprime: Single Cell\
+      \ 5'\n  - SC5P-PE: Paired-end Single Cell 5'\n  - SC5P-PE-v3: Paired-end Single\
+      \ Cell 5' v3\n  - SC5P-R2: R2-only Single Cell 5'\n  - SC5P-R2-v3: R2-only Single\
+      \ Cell 5' v3\n  - SCP5-PE-v3: Single Cell 5' paired-end v3 (GEM-X)\n  - SC5PHT\
+      \ : Single Cell 5' v2 HT\n  - SFRP: Fixed RNA Profiling (Singleplex)\n  - MFRP:\
+      \ Fixed RNA Profiling (Multiplex, Probe Barcode on R2)\n  - MFRP-R1: Fixed RNA\
+      \ Profiling (Multiplex, Probe Barcode on R1)\n  - MFRP-RNA: Fixed RNA Profiling\
+      \ (Multiplex, RNA, Probe Barcode on R2)\n  - MFRP-Ab: Fixed RNA Profiling (Multiplex,\
+      \ Antibody, Probe Barcode at R2:69)\n  - MFRP-Ab-R2pos50: Fixed RNA Profiling\
+      \ (Multiplex, Antibody, Probe Barcode at R2:50)\n  - MFRP-RNA-R1: Fixed RNA\
+      \ Profiling (Multiplex, RNA, Probe Barcode on R1)\n  - MFRP-Ab-R1: Fixed RNA\
+      \ Profiling (Multiplex, Antibody, Probe Barcode on R1)\n  - ARC-v1 for analyzing\
+      \ the Gene Expression portion of Multiome data. If Cell Ranger auto-detects\
+      \ ARC-v1 chemistry, an error is triggered.\nSee https://kb.10xgenomics.com/hc/en-us/articles/115003764132-How-does-Cell-Ranger-auto-detect-chemistry-\
+      \ for more information.\n"
+    info: null
+    default:
+    - "auto"
+    required: false
+    choices:
+    - "auto"
+    - "threeprime"
+    - "fiveprime"
+    - "SC3Pv1"
+    - "SC3Pv2"
+    - "SC3Pv3"
+    - "SC3Pv3-polyA"
+    - "SC3Pv4"
+    - "SC3Pv4-polyA"
+    - "SC3Pv3LT"
+    - "SC3Pv3HT"
+    - "SC3Pv3HT-polyA"
+    - "SC5P-PE"
+    - "SC5P-PE-v3"
+    - "SC5P-R2"
+    - "SC-FB"
+    - "SC5P-R2-v3"
+    - "SCP5-PE-v3"
+    - "SC5PHT"
+    - "MFRP"
+    - "MFRP-R1"
+    - "MFRP-RNA"
+    - "MFRP-Ab"
+    - "SFRP"
+    - "MFRP-Ab-R2pos50"
+    - "MFRP-RNA-R1"
+    - "MFRP-Ab-R1"
+    - "ARC-v1"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "VDJ related parameters"
+  arguments:
+  - type: "file"
+    name: "--vdj_reference"
+    description: "VDJ refence index built by Cell Ranger mkref."
+    info: null
+    example:
+    - "reference_vdj.tar.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--vdj_inner_enrichment_primers"
+    description: "V(D)J Immune Profiling libraries: if inner enrichment primers other\
+      \ than those provided \nin the 10x Genomics kits are used, they need to be specified\
+      \ here as a\ntext file with one primer per line.\n"
+    info: null
+    example:
+    - "enrichment_primers.txt"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--vdj_r1_length"
+    description: "Limit the length of the input Read 1 sequence of V(D)J libraries\
+      \ to the first N bases, where N is the user-supplied value.\nNote that the length\
+      \ includes the Barcode and UMI sequences so do not set this below 26.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--vdj_r2_length"
+    description: "Limit the length of the input Read 2 sequence of V(D)J libraries\
+      \ to the first N bases, where N is a user-supplied value. \nTrimming occurs\
+      \ before sequencing metrics are computed and therefore, limiting the length\
+      \ of Read 2 may affect Q30 scores\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--vdj_denovo"
+    description: "Run in reference-free mode (i.e., do not use annotations). This\
+      \ option is not supported for multiplexed experiments.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "3' Cell multiplexing parameters (CellPlex Multiplexing)"
+  arguments:
+  - type: "string"
+    name: "--cell_multiplex_oligo_ids"
+    alternatives:
+    - "--cmo_ids"
+    description: "The Cell Multiplexing oligo IDs used to multiplex this sample. If\
+      \ multiple CMOs were used for a sample,\nseparate IDs with a pipe (e.g., CMO301|CMO302).\
+      \ Required for Cell Multiplexing libraries.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "double"
+    name: "--min_assignment_confidence"
+    description: "The minimum estimated likelihood to call a sample as tagged with\
+      \ a Cell Multiplexing Oligo (CMO) instead of \"Unassigned\".\nUsers may wish\
+      \ to tolerate a higher rate of mis-assignment in order to obtain more singlets\
+      \ to include in their analysis,\nor a lower rate of mis-assignment at the cost\
+      \ of obtaining fewer singlets.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--cmo_set"
+    description: "Path to a custom CMO set CSV file, declaring CMO constructs and\
+      \ associated barcodes. If the default CMO reference IDs that are built into\n\
+      the Cell Ranger software are required, this option does not need to be used.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--barcode_sample_assignment"
+    description: "Path to a barcode-sample assignment CSV file that specifies the\
+      \ barcodes that belong to each sample.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Hashtag multiplexing parameters"
+  arguments:
+  - type: "string"
+    name: "--hashtag_ids"
+    description: "The hashtag IDs used to multiplex this sample. If multiple antibody\
+      \ hashtags were used for the same sample,\nyou can separate IDs with a pipe.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "On-chip multiplexing parameters"
+  arguments:
+  - type: "string"
+    name: "--ocm_barcode_ids"
+    description: "The OCM barcode IDs used to multiplex this sample. Must be one of\
+      \ OB1, OB2, OB3, OB4.\nIf multiple OCM Barcodes were used for the same sample,\
+      \ you can separate IDs\nwith a pipe (e.g., OB1|OB2).\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Flex multiplexing paramaters"
+  arguments:
+  - type: "file"
+    name: "--probe_set"
+    description: "A probe set reference CSV file. It specifies the sequences used\
+      \ as a reference for probe alignment and the gene ID associated with each probe.\n\
+      It must include 4 columns (probe file format 1.0.0): gene_id,probe_seq,probe_id,included,region\
+      \ and an optional 5th column (probe file format 1.0.1).\n- gene_id: The Ensembl\
+      \ gene identifier targeted by the probe.\n- probe_seq: The nucleotide sequence\
+      \ of the probe, which is complementary to the transcript sequence.\n- probe_id:\
+      \ The probe identifier, whose format is described in Probe identifiers.\n- included:\
+      \ A TRUE or FALSE flag specifying whether the probe is included in the filtered\
+      \ counts matrix output or excluded by the probe filter. \n            See filter-probes\
+      \ option of cellranger multi. All probes of a gene must be marked TRUE in the\
+      \ included column for that gene to be included.\n- region: Present only in v1.0.1\
+      \ probe set reference CSV. The gene boundary targeted by the probe. Accepted\
+      \ values are spliced or unspliced.\n\nThe file also contains a number of required\
+      \ metadata fields in the header in the format #key=value:\n- panel_name: The\
+      \ name of the probe set.\n- panel_type: Always predesigned for predesigned probe\
+      \ sets.\n- reference_genome: The reference genome build used for probe design.\n\
+      - reference_version: The version of the Cell Ranger reference transcriptome\
+      \ used for probe design.\n- probe_set_file_format: The version of the probe\
+      \ set file format specification that this file conforms to.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--filter_probes"
+    description: "If 'false', include all non-deprecated probes listed in the probe\
+      \ set reference CSV file.\nIf 'true' or not set, probes that are predicted to\
+      \ have off-target activity to homologous genes are excluded from analysis.\n\
+      Not filtering will result in UMI counts from all non-deprecated probes,\nincluding\
+      \ those with predicted off-target activity, to be used in the analysis.\nProbes\
+      \ whose ID is prefixed with DEPRECATED are always excluded from the analysis.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--probe_barcode_ids"
+    description: "The Fixed RNA Probe Barcode ID used for this sample, and for multiplex\
+      \ GEX + Antibody Capture libraries,\nthe corresponding Antibody Multiplexing\
+      \ Barcode IDs. 10x recommends specifying both barcodes (e.g., BC001+AB001)\n\
+      when an Antibody Capture library is present. The barcode pair order is BC+AB\
+      \ and they\nare separated with a \"+\" (no spaces). Alternatively, you can specify\
+      \ the Probe Barcode ID alone and\nCell Ranger's barcode pairing auto-detection\
+      \ algorithm will automatically match to the corresponding Antibody\nMultiplexing\
+      \ Barcode.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--emptydrops_minimum_umis"
+    description: "For singleplex Flex experiments, use this option to adjust the UMI\
+      \ cutoff during the second step of cell calling.\nCell Ranger will still perform\
+      \ the full cell calling process but will only evaluate barcodes with UMIs above\n\
+      the threshold you specify.\n"
+    info: null
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Antigen Capture (BEAM) libary arguments"
+  description: "These arguments are recommended if an Antigen Capture (BEAM) library\
+    \ is present. \nIt is needed to calculate the antigen specificity score.\n"
+  arguments:
+  - type: "string"
+    name: "--control_id"
+    description: "A user-defined ID for any negative controls used in the T/BCR Antigen\
+      \ Capture assay. Must match id specified in the feature reference CSV.\nMay\
+      \ only include ASCII characters and must not use whitespace, slash, quote, or\
+      \ comma characters. \nEach ID must be unique and must not collide with a gene\
+      \ identifier from the transcriptome.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--mhc_allele"
+    description: "The MHC allele for TCR Antigen Capture libraries. Must match mhc_allele\
+      \ name specified in the Feature Reference CSV.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "General arguments"
+  description: "These arguments are applicable to all library types.\n"
+  arguments:
+  - type: "boolean"
+    name: "--check_library_compatibility"
+    description: "Optional. This option allows users to disable the check that evaluates\
+      \ 10x Barcode overlap between\nibraries when multiple libraries are specified\
+      \ (e.g., Gene Expression + Antibody Capture). Setting\nthis option to false\
+      \ will disable the check across all library combinations. We recommend running\n\
+      this check (default), however if the pipeline errors out, users can bypass the\
+      \ check to generate\noutputs for troubleshooting.\n"
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output_raw"
+    description: "The raw output folder."
+    info: null
+    example:
+    - "output_dir"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_h5mu"
+    description: "Locations for the output files. Must contain a wildcard (*) character,\n\
+      which will be replaced with the sample name.\n"
+    info: null
+    example:
+    - "*.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--uns_metrics"
+    description: "Name of the .uns slot under which to QC metrics (if any)."
+    info: null
+    default:
+    - "metrics_cellranger"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "A pipeline for running Cell Ranger multi."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf2"
+- type: "file"
+  path: "10x_5k_anticmv/raw/"
+  dest: "10x_5k_anticmv/raw/"
+- type: "file"
+  path: "10x_5k_fixed/raw/"
+  dest: "10x_5k_fixed/raw/"
+- type: "file"
+  path: "reference_gencodev41_chr1"
+info:
+  name: "Cell Ranger multi"
+  test_dependencies:
+  - name: "cellranger_multi_test"
+    namespace: "test_workflows/ingestion"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "mapping/cellranger_multi"
+  alias: "cellranger_multi_component"
+  repository:
+    type: "local"
+- name: "convert/from_cellranger_multi_to_h5mu"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/ingestion/cellranger_multi/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/ingestion/cellranger_multi"
+  executable: "target/nextflow/workflows/ingestion/cellranger_multi/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/mapping/cellranger_multi"
+  - "target/nextflow/convert/from_cellranger_multi_to_h5mu"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/main.nf
new file mode 100644
index 0000000..7a67c35
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/main.nf
@@ -0,0 +1,4328 @@
+// cellranger_multi v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "cellranger_multi",
+  "namespace" : "workflows/ingestion",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Input files",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "The FASTQ files to be analyzed. FASTQ files should conform to the naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Feature type-specific input files",
+      "description" : "Helper functionality to allow feature type-specific input files, without the need to specify\nlibrary_type or library_id. The library_id will be inferred from the input paths.\n",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--gex_input",
+          "description" : "The FASTQ files to be analyzed for Gene Expression. FASTQ files should conform to the\nnaming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--abc_input",
+          "description" : "The FASTQ files to be analyzed for Antibody Capture. FASTQ files should conform to \nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--cgc_input",
+          "description" : "The FASTQ files to be analyzed for CRISPR Guide Capture. FASTQ files should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--mux_input",
+          "description" : "The FASTQ files to be analyzed for Multiplexing Capture. FASTQ files should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--vdj_input",
+          "description" : "The FASTQ files to be analyzed for VDJ. FASTQ files should conform to the\nnaming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--vdj_t_input",
+          "description" : "The FASTQ files to be analyzed for VDJ-T. FASTQ files should conform to the naming\nconventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--vdj_t_gd_input",
+          "description" : "The FASTQ files to be analyzed for VDJ-T-GD. FASTQ files should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--vdj_b_input",
+          "description" : "The FASTQ files to be analyzed for VDJ-B. FASTQ files should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--agc_input",
+          "description" : "The FASTQ files to be analyzed for Antigen Capture. FASTQ files should conform to\nthe naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Library arguments",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--library_id",
+          "description" : "The Illumina sample name to analyze. This must exactly match the 'Sample Name'part\nof the FASTQ files specified in the `--input` argument.\n",
+          "example" : [
+            "mysample1"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--library_type",
+          "description" : "The underlying feature type of the library.\n",
+          "example" : [
+            "Gene Expression"
+          ],
+          "required" : false,
+          "choices" : [
+            "Gene Expression",
+            "VDJ",
+            "VDJ-T",
+            "VDJ-B",
+            "VDJ-T-GD",
+            "Antibody Capture",
+            "CRISPR Guide Capture",
+            "Multiplexing Capture",
+            "Antigen Capture",
+            "Custom"
+          ],
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--library_subsample",
+          "description" : "The rate at which reads from the provided FASTQ files are sampled.\nMust be strictly greater than 0 and less than or equal to 1.\n",
+          "example" : [
+            "0.5"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--library_lanes",
+          "description" : "Lanes associated with this sample. Defaults to using all lanes.",
+          "example" : [
+            "1-4"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--library_chemistry",
+          "description" : "Only applicable to FRP. Library-specific assay configuration. By default,\nthe assay configuration is detected automatically. Typically, users will\nnot need to specify a chemistry.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Sample parameters",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--sample_ids",
+          "alternatives" : [
+            "--cell_multiplex_sample_id"
+          ],
+          "description" : "A name to identify a multiplexed sample. Must be alphanumeric with hyphens and/or underscores,\nand less than 64 characters. Required for Cell Multiplexing libraries.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--sample_description",
+          "alternatives" : [
+            "--cell_multiplex_description"
+          ],
+          "description" : "A description for the sample.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--sample_expect_cells",
+          "description" : "Expected number of recovered cells, used as input to cell calling algorithm.\n",
+          "example" : [
+            3000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--sample_force_cells",
+          "description" : "Force pipeline to use this number of cells, bypassing cell detection.\n",
+          "example" : [
+            3000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Feature Barcode library specific arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--feature_reference",
+          "description" : "Path to the Feature reference CSV file, declaring Feature Barcode constructs and associated barcodes.\nRequired only for Antibody Capture or CRISPR Guide Capture libraries.\nSee https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/using/feature-bc-analysis#feature-ref for more information.\\"\n",
+          "example" : [
+            "feature_reference.csv"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--feature_r1_length",
+          "description" : "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,\nwhere N is the user-supplied value. Note that the length includes the Barcode and UMI\nsequences so do not set this below 26.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--feature_r2_length",
+          "description" : "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,\nwhere N is a user-supplied value. Trimming occurs before sequencing metrics are computed\nand therefore, limiting the length of Read 2 may affect Q30 scores.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_crispr_umi",
+          "description" : "Set the minimum number of CRISPR guide RNA UMIs required for protospacer detection.\nIf a lower or higher sensitivity is desired for detection, this value can be customized\naccording to specific experimental needs. Applicable only to datasets that include a\nCRISPR Guide Capture library.\n",
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Gene expression arguments",
+      "description" : "Arguments relevant to the analysis of gene expression data.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--gex_reference",
+          "description" : "Genome refence index built by Cell Ranger mkref.",
+          "example" : [
+            "reference_genome.tar.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--gex_secondary_analysis",
+          "description" : "Whether or not to run the secondary analysis e.g. clustering.",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--gex_generate_bam",
+          "description" : "Whether to generate a BAM file.",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--tenx_cloud_token_path",
+          "description" : "The 10x Cloud Analysis user token used to enable cell annotation.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--cell_annotation_model",
+          "description" : "\\"Cell annotation model to use. If auto, uses the default model for the species.\nIf not given, does not run cell annotation.\\"\n",
+          "required" : false,
+          "choices" : [
+            "auto",
+            "human_pca_v1_beta",
+            "mouse_pca_v1_beta"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gex_expect_cells",
+          "description" : "Expected number of recovered cells, used as input to cell calling algorithm.\n",
+          "example" : [
+            3000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gex_force_cells",
+          "description" : "Force pipeline to use this number of cells, bypassing cell detection.\n",
+          "example" : [
+            3000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--gex_include_introns",
+          "description" : "Whether or not to include intronic reads in counts.\nThis option does not apply to Fixed RNA Profiling analysis.\n",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gex_r1_length",
+          "description" : "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,\nwhere N is the user-supplied value. Note that the length includes the Barcode and UMI\nsequences so do not set this below 26.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gex_r2_length",
+          "description" : "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,\nwhere N is a user-supplied value. Trimming occurs before sequencing metrics are computed\nand therefore, limiting the length of Read 2 may affect Q30 scores.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--gex_chemistry",
+          "description" : "Assay configuration. Either specify a single value which will be applied to all libraries,\nor a number of values that is equal to the number of libararies. The latter is only applicable\nto only applicable to Fixed RNA Profiling.\n  - auto: Chemistry autodetection (default)\n  - threeprime: Single Cell 3'\n  - SC3Pv1, SC3Pv2, SC3Pv3(-polyA), SC3Pv4(-polyA): Single Cell 3' v1, v2, v3, or v4\n  - SC3Pv3HT(-polyA): Single Cell 3' v3.1 HT\n  - SC-FB: Single Cell Antibody-only 3' v2 or 5'\n  - fiveprime: Single Cell 5'\n  - SC5P-PE: Paired-end Single Cell 5'\n  - SC5P-PE-v3: Paired-end Single Cell 5' v3\n  - SC5P-R2: R2-only Single Cell 5'\n  - SC5P-R2-v3: R2-only Single Cell 5' v3\n  - SCP5-PE-v3: Single Cell 5' paired-end v3 (GEM-X)\n  - SC5PHT : Single Cell 5' v2 HT\n  - SFRP: Fixed RNA Profiling (Singleplex)\n  - MFRP: Fixed RNA Profiling (Multiplex, Probe Barcode on R2)\n  - MFRP-R1: Fixed RNA Profiling (Multiplex, Probe Barcode on R1)\n  - MFRP-RNA: Fixed RNA Profiling (Multiplex, RNA, Probe Barcode on R2)\n  - MFRP-Ab: Fixed RNA Profiling (Multiplex, Antibody, Probe Barcode at R2:69)\n  - MFRP-Ab-R2pos50: Fixed RNA Profiling (Multiplex, Antibody, Probe Barcode at R2:50)\n  - MFRP-RNA-R1: Fixed RNA Profiling (Multiplex, RNA, Probe Barcode on R1)\n  - MFRP-Ab-R1: Fixed RNA Profiling (Multiplex, Antibody, Probe Barcode on R1)\n  - ARC-v1 for analyzing the Gene Expression portion of Multiome data. If Cell Ranger auto-detects ARC-v1 chemistry, an error is triggered.\nSee https://kb.10xgenomics.com/hc/en-us/articles/115003764132-How-does-Cell-Ranger-auto-detect-chemistry- for more information.\n",
+          "default" : [
+            "auto"
+          ],
+          "required" : false,
+          "choices" : [
+            "auto",
+            "threeprime",
+            "fiveprime",
+            "SC3Pv1",
+            "SC3Pv2",
+            "SC3Pv3",
+            "SC3Pv3-polyA",
+            "SC3Pv4",
+            "SC3Pv4-polyA",
+            "SC3Pv3LT",
+            "SC3Pv3HT",
+            "SC3Pv3HT-polyA",
+            "SC5P-PE",
+            "SC5P-PE-v3",
+            "SC5P-R2",
+            "SC-FB",
+            "SC5P-R2-v3",
+            "SCP5-PE-v3",
+            "SC5PHT",
+            "MFRP",
+            "MFRP-R1",
+            "MFRP-RNA",
+            "MFRP-Ab",
+            "SFRP",
+            "MFRP-Ab-R2pos50",
+            "MFRP-RNA-R1",
+            "MFRP-Ab-R1",
+            "ARC-v1"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "VDJ related parameters",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--vdj_reference",
+          "description" : "VDJ refence index built by Cell Ranger mkref.",
+          "example" : [
+            "reference_vdj.tar.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--vdj_inner_enrichment_primers",
+          "description" : "V(D)J Immune Profiling libraries: if inner enrichment primers other than those provided \nin the 10x Genomics kits are used, they need to be specified here as a\ntext file with one primer per line.\n",
+          "example" : [
+            "enrichment_primers.txt"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--vdj_r1_length",
+          "description" : "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases, where N is the user-supplied value.\nNote that the length includes the Barcode and UMI sequences so do not set this below 26.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--vdj_r2_length",
+          "description" : "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases, where N is a user-supplied value. \nTrimming occurs before sequencing metrics are computed and therefore, limiting the length of Read 2 may affect Q30 scores\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--vdj_denovo",
+          "description" : "Run in reference-free mode (i.e., do not use annotations). This option is not supported for multiplexed experiments.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "3' Cell multiplexing parameters (CellPlex Multiplexing)",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--cell_multiplex_oligo_ids",
+          "alternatives" : [
+            "--cmo_ids"
+          ],
+          "description" : "The Cell Multiplexing oligo IDs used to multiplex this sample. If multiple CMOs were used for a sample,\nseparate IDs with a pipe (e.g., CMO301|CMO302). Required for Cell Multiplexing libraries.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--min_assignment_confidence",
+          "description" : "The minimum estimated likelihood to call a sample as tagged with a Cell Multiplexing Oligo (CMO) instead of \\"Unassigned\\".\nUsers may wish to tolerate a higher rate of mis-assignment in order to obtain more singlets to include in their analysis,\nor a lower rate of mis-assignment at the cost of obtaining fewer singlets.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--cmo_set",
+          "description" : "Path to a custom CMO set CSV file, declaring CMO constructs and associated barcodes. If the default CMO reference IDs that are built into\nthe Cell Ranger software are required, this option does not need to be used.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--barcode_sample_assignment",
+          "description" : "Path to a barcode-sample assignment CSV file that specifies the barcodes that belong to each sample.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Hashtag multiplexing parameters",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--hashtag_ids",
+          "description" : "The hashtag IDs used to multiplex this sample. If multiple antibody hashtags were used for the same sample,\nyou can separate IDs with a pipe.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "On-chip multiplexing parameters",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--ocm_barcode_ids",
+          "description" : "The OCM barcode IDs used to multiplex this sample. Must be one of OB1, OB2, OB3, OB4.\nIf multiple OCM Barcodes were used for the same sample, you can separate IDs\nwith a pipe (e.g., OB1|OB2).\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Flex multiplexing paramaters",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--probe_set",
+          "description" : "A probe set reference CSV file. It specifies the sequences used as a reference for probe alignment and the gene ID associated with each probe.\nIt must include 4 columns (probe file format 1.0.0): gene_id,probe_seq,probe_id,included,region and an optional 5th column (probe file format 1.0.1).\n- gene_id: The Ensembl gene identifier targeted by the probe.\n- probe_seq: The nucleotide sequence of the probe, which is complementary to the transcript sequence.\n- probe_id: The probe identifier, whose format is described in Probe identifiers.\n- included: A TRUE or FALSE flag specifying whether the probe is included in the filtered counts matrix output or excluded by the probe filter. \n            See filter-probes option of cellranger multi. All probes of a gene must be marked TRUE in the included column for that gene to be included.\n- region: Present only in v1.0.1 probe set reference CSV. The gene boundary targeted by the probe. Accepted values are spliced or unspliced.\n\nThe file also contains a number of required metadata fields in the header in the format #key=value:\n- panel_name: The name of the probe set.\n- panel_type: Always predesigned for predesigned probe sets.\n- reference_genome: The reference genome build used for probe design.\n- reference_version: The version of the Cell Ranger reference transcriptome used for probe design.\n- probe_set_file_format: The version of the probe set file format specification that this file conforms to.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--filter_probes",
+          "description" : "If 'false', include all non-deprecated probes listed in the probe set reference CSV file.\nIf 'true' or not set, probes that are predicted to have off-target activity to homologous genes are excluded from analysis.\nNot filtering will result in UMI counts from all non-deprecated probes,\nincluding those with predicted off-target activity, to be used in the analysis.\nProbes whose ID is prefixed with DEPRECATED are always excluded from the analysis.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--probe_barcode_ids",
+          "description" : "The Fixed RNA Probe Barcode ID used for this sample, and for multiplex GEX + Antibody Capture libraries,\nthe corresponding Antibody Multiplexing Barcode IDs. 10x recommends specifying both barcodes (e.g., BC001+AB001)\nwhen an Antibody Capture library is present. The barcode pair order is BC+AB and they\nare separated with a \\"+\\" (no spaces). Alternatively, you can specify the Probe Barcode ID alone and\nCell Ranger's barcode pairing auto-detection algorithm will automatically match to the corresponding Antibody\nMultiplexing Barcode.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--emptydrops_minimum_umis",
+          "description" : "For singleplex Flex experiments, use this option to adjust the UMI cutoff during the second step of cell calling.\nCell Ranger will still perform the full cell calling process but will only evaluate barcodes with UMIs above\nthe threshold you specify.\n",
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Antigen Capture (BEAM) libary arguments",
+      "description" : "These arguments are recommended if an Antigen Capture (BEAM) library is present. \nIt is needed to calculate the antigen specificity score.\n",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--control_id",
+          "description" : "A user-defined ID for any negative controls used in the T/BCR Antigen Capture assay. Must match id specified in the feature reference CSV.\nMay only include ASCII characters and must not use whitespace, slash, quote, or comma characters. \nEach ID must be unique and must not collide with a gene identifier from the transcriptome.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--mhc_allele",
+          "description" : "The MHC allele for TCR Antigen Capture libraries. Must match mhc_allele name specified in the Feature Reference CSV.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "General arguments",
+      "description" : "These arguments are applicable to all library types.\n",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--check_library_compatibility",
+          "description" : "Optional. This option allows users to disable the check that evaluates 10x Barcode overlap between\nibraries when multiple libraries are specified (e.g., Gene Expression + Antibody Capture). Setting\nthis option to false will disable the check across all library combinations. We recommend running\nthis check (default), however if the pipeline errors out, users can bypass the check to generate\noutputs for troubleshooting.\n",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output_raw",
+          "description" : "The raw output folder.",
+          "example" : [
+            "output_dir"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_h5mu",
+          "description" : "Locations for the output files. Must contain a wildcard (*) character,\nwhich will be replaced with the sample name.\n",
+          "example" : [
+            "*.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--uns_metrics",
+          "description" : "Name of the .uns slot under which to QC metrics (if any).",
+          "default" : [
+            "metrics_cellranger"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "A pipeline for running Cell Ranger multi.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf2"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_anticmv/raw/",
+      "dest" : "10x_5k_anticmv/raw/"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_fixed/raw/",
+      "dest" : "10x_5k_fixed/raw/"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/reference_gencodev41_chr1"
+    }
+  ],
+  "info" : {
+    "name" : "Cell Ranger multi",
+    "test_dependencies" : [
+      {
+        "name" : "cellranger_multi_test",
+        "namespace" : "test_workflows/ingestion"
+      }
+    ]
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "mapping/cellranger_multi",
+      "alias" : "cellranger_multi_component",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "convert/from_cellranger_multi_to_h5mu",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/ingestion/cellranger_multi/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/ingestion/cellranger_multi",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { cellranger_multi as cellranger_multi_component_viashalias } from "${meta.resources_dir}/../../../../nextflow/mapping/cellranger_multi/main.nf"
+cellranger_multi_component = cellranger_multi_component_viashalias.run(key: "cellranger_multi_component")
+include { from_cellranger_multi_to_h5mu } from "${meta.resources_dir}/../../../../nextflow/convert/from_cellranger_multi_to_h5mu/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+process cellrangerMultiStub {
+  input:
+    tuple val(id), val(unused)
+
+  output:
+    tuple val(id), path("raw_dir"), path("*.h5mu"), path("samples.csv")
+
+  script:
+    """
+    echo "This process is not meant to be run without -stub being defined."
+    exit 1
+    """
+
+  stub:
+    """
+    mkdir raw_dir
+    touch ${id}_sample_1.h5mu
+    touch ${id}_sample_2.h5mu
+    touch ${id}_sample_3.h5mu
+    echo -e "sample_name,file\n${id}_sample_1,${id}_sample_1.h5mu\n${id}_sample_2,${id}_sample_2.h5mu\n${id}_sample_3,${id}_sample_3.h5mu" > samples.csv
+    """
+}
+
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+  cellranger_ch = input_ch
+    | cellranger_multi_component.run(
+      fromState: [
+        "input": "input",
+        "gex_input": "gex_input",
+        "abc_input": "abc_input",
+        "cgc_input": "cgc_input",
+        "mux_input": "mux_input",
+        "vdj_input": "vdj_input",
+        "vdj_t_input": "vdj_t_input",
+        "vdj_t_gd_input": "vdj_t_gd_input",
+        "vdj_b_input": "vdj_b_input",
+        "vdj_denovo": "vdj_denovo",
+        "agc_input": "agc_input",
+        "library_id": "library_id",
+        "library_type": "library_type",
+        "library_subsample": "library_subsample",
+        "library_lanes": "library_lanes",
+        "library_chemistry": "library_chemistry",
+        "sample_ids": "sample_ids",
+        "sample_description": "sample_description",
+        "sample_expect_cells": "sample_expect_cells",
+        "sample_force_cells": "sample_force_cells",
+        "feature_reference": "feature_reference",
+        "feature_r1_length": "feature_r1_length",
+        "feature_r2_length": "feature_r2_length",
+        "gex_reference": "gex_reference",
+        "gex_secondary_analysis": "gex_secondary_analysis",
+        "gex_generate_bam": "gex_generate_bam",
+        "gex_expect_cells": "gex_expect_cells",
+        "gex_force_cells": "gex_force_cells",
+        "gex_include_introns": "gex_include_introns",
+        "gex_r1_length": "gex_r1_length",
+        "gex_r2_length": "gex_r2_length",
+        "gex_chemistry": "gex_chemistry",
+        "vdj_reference": "vdj_reference",
+        "vdj_inner_enrichment_primers": "vdj_inner_enrichment_primers",
+        "vdj_r1_length": "vdj_r1_length",
+        "vdj_r2_length": "vdj_r2_length",
+        "cell_multiplex_oligo_ids": "cell_multiplex_oligo_ids",
+        "min_assignment_confidence": "min_assignment_confidence",
+        "cmo_set": "cmo_set",
+        "barcode_sample_assignment": "barcode_sample_assignment",
+        "ocm_barcode_ids": "ocm_barcode_ids",
+        "min_crispr_umi": "min_crispr_umi",
+        "emptydrops_minimum_umis": "emptydrops_minimum_umis",
+        "hashtag_ids": "hashtag_ids",
+        "probe_set": "probe_set",
+        "filter_probes": "filter_probes",
+        "probe_barcode_ids": "probe_barcode_ids",
+        "control_id": "control_id",
+        "mhc_allele": "mhc_allele",
+        "check_library_compatibility": "check_library_compatibility",
+        "output": "output_raw",
+      ],
+      toState: [
+        "output_raw": "output", 
+        "input": "output"
+      ]
+    )
+
+  h5mu_ch = cellranger_ch
+    | from_cellranger_multi_to_h5mu.run(
+      fromState: {id, state ->
+        [
+          "input": state.input,
+          "uns_metrics": state.uns_metrics,
+          "output_compression": "gzip"
+        ]
+      },
+      toState: {id, output, state -> 
+        [
+          "sample_csv": output.sample_csv,
+          "output_h5mu": output.output,
+          "output_raw": state.output_raw
+        ]
+      },
+      filter: {!workflow.stubRun},
+    )
+ 
+  h5mu_stub_ch = cellranger_ch
+    | filter{workflow.stubRun}
+    | map {id, state -> [id, state.input]}
+    | cellrangerMultiStub
+    | map {id, raw_dir, output, output_files ->
+      [id, ["output_raw": raw_dir, "output_h5mu": output, "sample_csv": output_files]]
+    }
+
+  output_ch = h5mu_stub_ch.concat(h5mu_ch)
+    | flatMap {id, state ->
+      def h5mu_list = state.output_h5mu
+      def csv = state.sample_csv.splitCsv(strip: true, sep: ",").findAll{!it[0].startsWith("#")}
+      def header = csv.head()
+      def samples = csv.tail().collect { row ->
+          [header, row].transpose().collectEntries()
+      }
+      println "Samples: $samples" 
+      def result = h5mu_list.collect{ h5mu_file ->
+        println "H5mu: ${h5mu_file}, getName: ${h5mu_file.getName()}"
+        def corresponding_csv_entry = samples.find{h5mu_file.getName() == it.file}
+        print "CSV entry: $corresponding_csv_entry"
+        // The cellranger component used to only be able to output a single h5mu file
+        // In cases where cell multiplexing is not used (1 output sample), it uses 'run' for the sample ID as a dummy.
+        // This sample ID 'run' was never used for the ID of the channel events.
+        // So here we overwrite this 'run' id with the name of the input event.
+        def new_id = h5mu_list.size() == 1 ? id : corresponding_csv_entry.sample_name
+        return [ new_id, ["output_h5mu": h5mu_file, "output_raw": state.output_raw, "_meta": ["join_id": id]]]
+      }
+      return result
+    }
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/nextflow.config
new file mode 100644
index 0000000..0d4710f
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/ingestion/cellranger_multi'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'A pipeline for running Cell Ranger multi.'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/nextflow_schema.json
new file mode 100644
index 0000000..61d9fb7
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/nextflow_schema.json
@@ -0,0 +1,576 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "cellranger_multi",
+  "description": "A pipeline for running Cell Ranger multi.",
+  "type": "object",
+  "$defs": {
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output_raw": {
+          "type": "string",
+          "format": "path",
+          "description": "The raw output folder.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output_raw\"`, direction: `output`, example: `\"output_dir\"`. ",
+          "default": "$id.$key.output_raw"
+        },
+        "output_h5mu": {
+          "type": "string",
+          "format": "path",
+          "description": "Locations for the output files",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output_h5mu.h5mu\"`, direction: `output`, example: `\"*.h5mu\"`. ",
+          "default": "$id.$key.output_h5mu.h5mu"
+        },
+        "uns_metrics": {
+          "type": "string",
+          "description": "Name of the .uns slot under which to QC metrics (if any).",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"metrics_cellranger\"`. ",
+          "default": "metrics_cellranger"
+        }
+      }
+    },
+    "input files": {
+      "title": "Input files",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        }
+      }
+    },
+    "feature type-specific input files": {
+      "title": "Feature type-specific input files",
+      "type": "object",
+      "description": "Helper functionality to allow feature type-specific input files, without the need to specify\nlibrary_type or library_id. The library_id will be inferred from the input paths.\n",
+      "properties": {
+        "gex_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for Gene Expression",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "abc_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for Antibody Capture",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "cgc_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for CRISPR Guide Capture",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "mux_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for Multiplexing Capture",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "vdj_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for VDJ",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "vdj_t_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for VDJ-T",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "vdj_t_gd_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for VDJ-T-GD",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "vdj_b_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for VDJ-B",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        },
+        "agc_input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed for Antigen Capture",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        }
+      }
+    },
+    "library arguments": {
+      "title": "Library arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "library_id": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The Illumina sample name to analyze",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"mysample1\"]`. "
+        },
+        "library_type": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The underlying feature type of the library.\n",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"Gene Expression\"]`, choices: ``Gene Expression`, `VDJ`, `VDJ-T`, `VDJ-B`, `VDJ-T-GD`, `Antibody Capture`, `CRISPR Guide Capture`, `Multiplexing Capture`, `Antigen Capture`, `Custom``. "
+        },
+        "library_subsample": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The rate at which reads from the provided FASTQ files are sampled.\nMust be strictly greater than 0 and less than or equal to 1.\n",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"0.5\"]`. "
+        },
+        "library_lanes": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Lanes associated with this sample",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"1-4\"]`. "
+        },
+        "library_chemistry": {
+          "type": "string",
+          "description": "Only applicable to FRP",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "sample parameters": {
+      "title": "Sample parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "sample_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "A name to identify a multiplexed sample",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "sample_description": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "A description for the sample.",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "sample_expect_cells": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Expected number of recovered cells, used as input to cell calling algorithm.\n",
+          "help_text": "Type: `integer`, multiple: `True`, example: `[3000]`. "
+        },
+        "sample_force_cells": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Force pipeline to use this number of cells, bypassing cell detection.\n",
+          "help_text": "Type: `integer`, multiple: `True`, example: `[3000]`. "
+        }
+      }
+    },
+    "feature barcode library specific arguments": {
+      "title": "Feature Barcode library specific arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "feature_reference": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to the Feature reference CSV file, declaring Feature Barcode constructs and associated barcodes.\nRequired only for Antibody Capture or CRISPR Guide Capture libraries.\nSee https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/using/feature-bc-analysis#feature-ref for more information.\"\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"feature_reference.csv\"`. "
+        },
+        "feature_r1_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,\nwhere N is the user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "feature_r2_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,\nwhere N is a user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "min_crispr_umi": {
+          "type": "integer",
+          "description": "Set the minimum number of CRISPR guide RNA UMIs required for protospacer detection.\nIf a lower or higher sensitivity is desired for detection, this value can be customized\naccording to specific experimental needs",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        }
+      }
+    },
+    "gene expression arguments": {
+      "title": "Gene expression arguments",
+      "type": "object",
+      "description": "Arguments relevant to the analysis of gene expression data.",
+      "properties": {
+        "gex_reference": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Genome refence index built by Cell Ranger mkref.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"reference_genome.tar.gz\"`. "
+        },
+        "gex_secondary_analysis": {
+          "type": "boolean",
+          "description": "Whether or not to run the secondary analysis e.g",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "gex_generate_bam": {
+          "type": "boolean",
+          "description": "Whether to generate a BAM file.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "tenx_cloud_token_path": {
+          "type": "string",
+          "format": "path",
+          "description": "The 10x Cloud Analysis user token used to enable cell annotation.",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`. "
+        },
+        "cell_annotation_model": {
+          "type": "string",
+          "description": "\"Cell annotation model to use",
+          "help_text": "Type: `string`, multiple: `False`, choices: ``auto`, `human_pca_v1_beta`, `mouse_pca_v1_beta``. ",
+          "enum": [
+            "auto",
+            "human_pca_v1_beta",
+            "mouse_pca_v1_beta"
+          ]
+        },
+        "gex_expect_cells": {
+          "type": "integer",
+          "description": "Expected number of recovered cells, used as input to cell calling algorithm.\n",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3000`. "
+        },
+        "gex_force_cells": {
+          "type": "integer",
+          "description": "Force pipeline to use this number of cells, bypassing cell detection.\n",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3000`. "
+        },
+        "gex_include_introns": {
+          "type": "boolean",
+          "description": "Whether or not to include intronic reads in counts.\nThis option does not apply to Fixed RNA Profiling analysis.\n",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "gex_r1_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,\nwhere N is the user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "gex_r2_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,\nwhere N is a user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "gex_chemistry": {
+          "type": "string",
+          "description": "Assay configuration",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"auto\"`, choices: ``auto`, `threeprime`, `fiveprime`, `SC3Pv1`, `SC3Pv2`, `SC3Pv3`, `SC3Pv3-polyA`, `SC3Pv4`, `SC3Pv4-polyA`, `SC3Pv3LT`, `SC3Pv3HT`, `SC3Pv3HT-polyA`, `SC5P-PE`, `SC5P-PE-v3`, `SC5P-R2`, `SC-FB`, `SC5P-R2-v3`, `SCP5-PE-v3`, `SC5PHT`, `MFRP`, `MFRP-R1`, `MFRP-RNA`, `MFRP-Ab`, `SFRP`, `MFRP-Ab-R2pos50`, `MFRP-RNA-R1`, `MFRP-Ab-R1`, `ARC-v1``. ",
+          "enum": [
+            "auto",
+            "threeprime",
+            "fiveprime",
+            "SC3Pv1",
+            "SC3Pv2",
+            "SC3Pv3",
+            "SC3Pv3-polyA",
+            "SC3Pv4",
+            "SC3Pv4-polyA",
+            "SC3Pv3LT",
+            "SC3Pv3HT",
+            "SC3Pv3HT-polyA",
+            "SC5P-PE",
+            "SC5P-PE-v3",
+            "SC5P-R2",
+            "SC-FB",
+            "SC5P-R2-v3",
+            "SCP5-PE-v3",
+            "SC5PHT",
+            "MFRP",
+            "MFRP-R1",
+            "MFRP-RNA",
+            "MFRP-Ab",
+            "SFRP",
+            "MFRP-Ab-R2pos50",
+            "MFRP-RNA-R1",
+            "MFRP-Ab-R1",
+            "ARC-v1"
+          ],
+          "default": "auto"
+        }
+      }
+    },
+    "vdj related parameters": {
+      "title": "VDJ related parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "vdj_reference": {
+          "type": "string",
+          "format": "path",
+          "description": "VDJ refence index built by Cell Ranger mkref.",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"reference_vdj.tar.gz\"`. "
+        },
+        "vdj_inner_enrichment_primers": {
+          "type": "string",
+          "format": "path",
+          "description": "V(D)J Immune Profiling libraries: if inner enrichment primers other than those provided \nin the 10x Genomics kits are used, they need to be specified here as a\ntext file with one primer per line.\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"enrichment_primers.txt\"`. "
+        },
+        "vdj_r1_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases, where N is the user-supplied value.\nNote that the length includes the Barcode and UMI sequences so do not set this below 26.\n",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "vdj_r2_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases, where N is a user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "vdj_denovo": {
+          "type": "boolean",
+          "description": "Run in reference-free mode (i.e., do not use annotations)",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        }
+      }
+    },
+    "3' cell multiplexing parameters (cellplex multiplexing)": {
+      "title": "3' Cell multiplexing parameters (CellPlex Multiplexing)",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "cell_multiplex_oligo_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The Cell Multiplexing oligo IDs used to multiplex this sample",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "min_assignment_confidence": {
+          "type": "number",
+          "description": "The minimum estimated likelihood to call a sample as tagged with a Cell Multiplexing Oligo (CMO) instead of \"Unassigned\".\nUsers may wish to tolerate a higher rate of mis-assignment in order to obtain more singlets to include in their analysis,\nor a lower rate of mis-assignment at the cost of obtaining fewer singlets.\n",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "cmo_set": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to a custom CMO set CSV file, declaring CMO constructs and associated barcodes",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`. "
+        },
+        "barcode_sample_assignment": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to a barcode-sample assignment CSV file that specifies the barcodes that belong to each sample.\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`. "
+        }
+      }
+    },
+    "hashtag multiplexing parameters": {
+      "title": "Hashtag multiplexing parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "hashtag_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The hashtag IDs used to multiplex this sample",
+          "help_text": "Type: `string`, multiple: `True`. "
+        }
+      }
+    },
+    "on-chip multiplexing parameters": {
+      "title": "On-chip multiplexing parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "ocm_barcode_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The OCM barcode IDs used to multiplex this sample",
+          "help_text": "Type: `string`, multiple: `True`. "
+        }
+      }
+    },
+    "flex multiplexing paramaters": {
+      "title": "Flex multiplexing paramaters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "probe_set": {
+          "type": "string",
+          "format": "path",
+          "description": "A probe set reference CSV file",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`. "
+        },
+        "filter_probes": {
+          "type": "boolean",
+          "description": "If 'false', include all non-deprecated probes listed in the probe set reference CSV file.\nIf 'true' or not set, probes that are predicted to have off-target activity to homologous genes are excluded from analysis.\nNot filtering will result in UMI counts from all non-deprecated probes,\nincluding those with predicted off-target activity, to be used in the analysis.\nProbes whose ID is prefixed with DEPRECATED are always excluded from the analysis.\n",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        },
+        "probe_barcode_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The Fixed RNA Probe Barcode ID used for this sample, and for multiplex GEX + Antibody Capture libraries,\nthe corresponding Antibody Multiplexing Barcode IDs",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "emptydrops_minimum_umis": {
+          "type": "integer",
+          "description": "For singleplex Flex experiments, use this option to adjust the UMI cutoff during the second step of cell calling.\nCell Ranger will still perform the full cell calling process but will only evaluate barcodes with UMIs above\nthe threshold you specify.\n",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        }
+      }
+    },
+    "antigen capture (beam) libary arguments": {
+      "title": "Antigen Capture (BEAM) libary arguments",
+      "type": "object",
+      "description": "These arguments are recommended if an Antigen Capture (BEAM) library is present. \nIt is needed to calculate the antigen specificity score.\n",
+      "properties": {
+        "control_id": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "A user-defined ID for any negative controls used in the T/BCR Antigen Capture assay",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "mhc_allele": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The MHC allele for TCR Antigen Capture libraries",
+          "help_text": "Type: `string`, multiple: `True`. "
+        }
+      }
+    },
+    "general arguments": {
+      "title": "General arguments",
+      "type": "object",
+      "description": "These arguments are applicable to all library types.\n",
+      "properties": {
+        "check_library_compatibility": {
+          "type": "boolean",
+          "description": "Optional",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/input files"
+    },
+    {
+      "$ref": "#/$defs/feature type-specific input files"
+    },
+    {
+      "$ref": "#/$defs/library arguments"
+    },
+    {
+      "$ref": "#/$defs/sample parameters"
+    },
+    {
+      "$ref": "#/$defs/feature barcode library specific arguments"
+    },
+    {
+      "$ref": "#/$defs/gene expression arguments"
+    },
+    {
+      "$ref": "#/$defs/vdj related parameters"
+    },
+    {
+      "$ref": "#/$defs/3' cell multiplexing parameters (cellplex multiplexing)"
+    },
+    {
+      "$ref": "#/$defs/hashtag multiplexing parameters"
+    },
+    {
+      "$ref": "#/$defs/on-chip multiplexing parameters"
+    },
+    {
+      "$ref": "#/$defs/flex multiplexing paramaters"
+    },
+    {
+      "$ref": "#/$defs/antigen capture (beam) libary arguments"
+    },
+    {
+      "$ref": "#/$defs/general arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/.config.vsh.yaml
new file mode 100644
index 0000000..937c8f2
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/.config.vsh.yaml
@@ -0,0 +1,417 @@
+name: "harmony_leiden"
+namespace: "workflows/integration"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Path to the sample. Mutually exclusive with the 'input_uri' argument."
+    info: null
+    example:
+    - "dataset.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--tiledb_input_uri"
+    description: "A URI containing the TileDB-SOMA objects. Mutually exclusive with\
+      \ the 'input' argument."
+    info: null
+    example:
+    - "s3://bucket/path"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--tiledb_s3_region"
+    description: "Region where the TileDB-SOMA database is hosted.\n"
+    info: null
+    example:
+    - "us-west-2"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--tiledb_endpoint"
+    description: "Custom endpoint to use to connect to S3\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--tiledb_s3_no_sign_request"
+    description: "Do not sign S3 requests. Credentials will not be loaded if this\
+      \ argument is provided.\n"
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Input slots"
+  arguments:
+  - type: "string"
+    name: "--layer"
+    description: "use specified layer for expression values instead of the .X object\
+      \ from the modality."
+    info: null
+    default:
+    - "log_normalized"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_tiledb"
+    description: "Output the TileDB database to the specified directory instead of\
+      \ adding it to the existing database.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Neighbour calculation"
+  arguments:
+  - type: "string"
+    name: "--uns_neighbors"
+    description: "In which .uns slot to store various neighbor output objects."
+    info: null
+    default:
+    - "harmonypy_integration_neighbors"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsp_neighbor_distances"
+    description: "In which .obsp slot to store the distance matrix between the resulting\
+      \ neighbors."
+    info: null
+    default:
+    - "harmonypy_integration_distances"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsp_neighbor_connectivities"
+    description: "In which .obsp slot to store the connectivities matrix between the\
+      \ resulting neighbors."
+    info: null
+    default:
+    - "harmonypy_integration_connectivities"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Harmony integration options"
+  arguments:
+  - type: "string"
+    name: "--embedding"
+    description: "Embedding to use as input"
+    info: null
+    default:
+    - "X_pca"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_integrated"
+    description: "In which .obsm slot to store the resulting integrated embedding."
+    info: null
+    default:
+    - "X_pca_integrated"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_covariates"
+    description: "The .obs field(s) that define the covariate(s) to regress out."
+    info: null
+    example:
+    - "batch"
+    - "sample"
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "double"
+    name: "--theta"
+    description: "Diversity clustering penalty parameter. Can be set as a single value\
+      \ for all batch observations or as multiple values, one for each observation\
+      \ in the batches defined by --obs_covariates. theta=0 does not encourage any\
+      \ diversity. Larger values of theta\nresult in more diverse clusters.\"\n"
+    info: null
+    default:
+    - 2.0
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Clustering options"
+  arguments:
+  - type: "string"
+    name: "--obs_cluster"
+    description: "Prefix for the .obs keys under which to add the cluster labels.\
+      \ Newly created columns in .obs will \nbe created from the specified value for\
+      \ '--obs_cluster' suffixed with an underscore and one of the resolutions\nresolutions\
+      \ specified in '--leiden_resolution'.\n"
+    info: null
+    default:
+    - "harmony_integration_leiden"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--leiden_resolution"
+    description: "Control the coarseness of the clustering. Higher values lead to\
+      \ more clusters."
+    info: null
+    default:
+    - 1.0
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Umap options"
+  arguments:
+  - type: "string"
+    name: "--obsm_umap"
+    description: "In which .obsm slot to store the resulting UMAP embedding."
+    info: null
+    default:
+    - "X_leiden_harmony_umap"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Run harmony integration followed by neighbour calculations, leiden clustering\
+  \ and run umap on the result."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_tiledb_wf"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf2"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "integrate/harmonypy"
+  repository:
+    type: "local"
+- name: "workflows/multiomics/neighbors_leiden_umap"
+  repository:
+    type: "local"
+- name: "convert/from_tiledb_to_h5mu"
+  repository:
+    type: "local"
+- name: "tiledb/move_mudata_obsm_to_tiledb"
+  repository:
+    type: "local"
+- name: "tiledb/move_mudata_obs_to_tiledb"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/integration/harmony_leiden/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/integration/harmony_leiden"
+  executable: "target/nextflow/workflows/integration/harmony_leiden/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/integrate/harmonypy"
+  - "target/nextflow/workflows/multiomics/neighbors_leiden_umap"
+  - "target/nextflow/convert/from_tiledb_to_h5mu"
+  - "target/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb"
+  - "target/_private/nextflow/tiledb/move_mudata_obs_to_tiledb"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/main.nf
new file mode 100644
index 0000000..2239359
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/main.nf
@@ -0,0 +1,3801 @@
+// harmony_leiden v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "harmony_leiden",
+  "namespace" : "workflows/integration",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Path to the sample. Mutually exclusive with the 'input_uri' argument.",
+          "example" : [
+            "dataset.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--tiledb_input_uri",
+          "description" : "A URI containing the TileDB-SOMA objects. Mutually exclusive with the 'input' argument.",
+          "example" : [
+            "s3://bucket/path"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--tiledb_s3_region",
+          "description" : "Region where the TileDB-SOMA database is hosted.\n",
+          "example" : [
+            "us-west-2"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--tiledb_endpoint",
+          "description" : "Custom endpoint to use to connect to S3\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--tiledb_s3_no_sign_request",
+          "description" : "Do not sign S3 requests. Credentials will not be loaded if this argument is provided.\n",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Input slots",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "use specified layer for expression values instead of the .X object from the modality.",
+          "default" : [
+            "log_normalized"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_tiledb",
+          "description" : "Output the TileDB database to the specified directory instead of adding it to the existing database.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Neighbour calculation",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--uns_neighbors",
+          "description" : "In which .uns slot to store various neighbor output objects.",
+          "default" : [
+            "harmonypy_integration_neighbors"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsp_neighbor_distances",
+          "description" : "In which .obsp slot to store the distance matrix between the resulting neighbors.",
+          "default" : [
+            "harmonypy_integration_distances"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsp_neighbor_connectivities",
+          "description" : "In which .obsp slot to store the connectivities matrix between the resulting neighbors.",
+          "default" : [
+            "harmonypy_integration_connectivities"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Harmony integration options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--embedding",
+          "description" : "Embedding to use as input",
+          "default" : [
+            "X_pca"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_integrated",
+          "description" : "In which .obsm slot to store the resulting integrated embedding.",
+          "default" : [
+            "X_pca_integrated"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_covariates",
+          "description" : "The .obs field(s) that define the covariate(s) to regress out.",
+          "example" : [
+            "batch",
+            "sample"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--theta",
+          "description" : "Diversity clustering penalty parameter. Can be set as a single value for all batch observations or as multiple values, one for each observation in the batches defined by --obs_covariates. theta=0 does not encourage any diversity. Larger values of theta\nresult in more diverse clusters.\\"\n",
+          "default" : [
+            2.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Clustering options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obs_cluster",
+          "description" : "Prefix for the .obs keys under which to add the cluster labels. Newly created columns in .obs will \nbe created from the specified value for '--obs_cluster' suffixed with an underscore and one of the resolutions\nresolutions specified in '--leiden_resolution'.\n",
+          "default" : [
+            "harmony_integration_leiden"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--leiden_resolution",
+          "description" : "Control the coarseness of the clustering. Higher values lead to more clusters.",
+          "default" : [
+            1.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Umap options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obsm_umap",
+          "description" : "In which .obsm slot to store the resulting UMAP embedding.",
+          "default" : [
+            "X_leiden_harmony_umap"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Run harmony integration followed by neighbour calculations, leiden clustering and run umap on the result.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_tiledb_wf"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf2"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "integrate/harmonypy",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/multiomics/neighbors_leiden_umap",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "convert/from_tiledb_to_h5mu",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "tiledb/move_mudata_obsm_to_tiledb",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "tiledb/move_mudata_obs_to_tiledb",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/integration/harmony_leiden/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/integration/harmony_leiden",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { harmonypy } from "${meta.resources_dir}/../../../../nextflow/integrate/harmonypy/main.nf"
+include { neighbors_leiden_umap } from "${meta.resources_dir}/../../../../nextflow/workflows/multiomics/neighbors_leiden_umap/main.nf"
+include { from_tiledb_to_h5mu } from "${meta.resources_dir}/../../../../nextflow/convert/from_tiledb_to_h5mu/main.nf"
+include { move_mudata_obsm_to_tiledb } from "${meta.resources_dir}/../../../../_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/main.nf"
+include { move_mudata_obs_to_tiledb } from "${meta.resources_dir}/../../../../_private/nextflow/tiledb/move_mudata_obs_to_tiledb/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+  output_ch = input_ch
+    | map {id, state ->
+      if (!workflow.stubRun) {
+        assert state.input || state.tiledb_input_uri: "Either --input or --tiledb_input_uri must be defined"
+        assert !(state.input && state.tiledb_input_uri): "Values were specified for both --input and --tiledb_input_uri. Please choose one."
+        assert state.tiledb_s3_region || !state.tiledb_input_uri: "Specifying 'tiledb_s3_region' also requires 'tiledb_input_uri'."
+        assert !state.tiledb_input_uri || state.layer: "When using tileDB input, you must specify a layer using --layer"
+      }
+      [id, state]
+    }
+    // Make sure there is not conflict between the output from this workflow
+    // And the output from any of the components
+    | map {id, state ->
+      def new_state = state + [
+        "workflow_output": state.output,
+        "output_tiledb": state.tiledb_input_uri ? state.output_tiledb : null
+      ]
+      [id, new_state]
+    }
+    | from_tiledb_to_h5mu.run(
+      runIf: {id, state -> state.tiledb_input_uri},
+      fromState: { id, state -> 
+        [
+          "input_uri": state.tiledb_input_uri,
+          "s3_region": state.tiledb_s3_region,
+          "endpoint": state.tiledb_endpoint,
+          "input_modality": state.modality,
+          "output_modality": state.modality,
+          "input_layers": [ state.layer ],
+          "s3_no_sign_request": state.tiledb_s3_no_sign_request
+        ]
+      },
+      toState: [
+        "input": "output",
+      ]
+    )
+    // run harmonypy
+    | harmonypy.run(
+      fromState: [
+          "input": "input",
+          "modality": "modality",
+          "obsm_input": "embedding",
+          "obs_covariates": "obs_covariates",
+          "obsm_output": "obsm_integrated",
+          "theta": "theta"
+      ],
+      toState: ["input": "output"]
+    )
+    | neighbors_leiden_umap.run(
+      fromState: [
+        "input": "input",
+        "modality": "modality",
+        "obsm_input": "obsm_integrated",
+        "output": "workflow_output",
+        "uns_neighbors": "uns_neighbors",
+        "obsp_neighbor_distances": "obsp_neighbor_distances",
+        "obsp_neighbor_connectivities": "obsp_neighbor_connectivities",
+        "leiden_resolution": "leiden_resolution",
+        "obs_cluster": "obs_cluster",
+        "obsm_umap": "obsm_umap",
+      ],
+      toState: ["output": "output"]
+    )
+    | move_mudata_obsm_to_tiledb.run(
+      runIf: {id, state -> state.tiledb_input_uri},
+      fromState: {id, state ->
+        [
+          "input_mudata": state.output,
+          "modality": state.modality,
+          "input_uri": state.tiledb_input_uri,
+          "s3_region": state.tiledb_s3_region,
+          "endpoint": state.tiledb_endpoint,
+          "output_modality": state.modality,
+          "obsm_input": [state.obsm_umap, state.obsm_integrated],
+          "output_tiledb": state.output_tiledb,
+          "s3_no_sign_request": state.tiledb_s3_no_sign_request
+        ]
+      },
+      toState: ["previous_output_tiledb": "output_tiledb"]
+    )
+    | move_mudata_obs_to_tiledb.run(
+      runIf: {id, state -> state.tiledb_input_uri},
+      fromState: {id, state ->
+        def new_state = [
+          "input_mudata": state.output,
+          "modality": state.modality,
+          "s3_region": state.tiledb_s3_region,
+          "endpoint": state.tiledb_endpoint,
+          "output_modality": state.modality,
+          "obs_input": state.leiden_resolution.collect{state.obs_cluster + "_" + it},
+          "output_tiledb": state.output_tiledb,
+          "s3_no_sign_request": state.tiledb_s3_no_sign_request
+        ]
+        if (state.previous_output_tiledb && file(state.previous_output_tiledb).exists()) {
+          new_state += ["input_dir": state.previous_output_tiledb]
+        } else {
+          new_state += ["input_uri": state.tiledb_input_uri]
+        }
+        return new_state
+      },
+      toState: ["output_tiledb": "output_tiledb"]
+    )
+    | setState(["output", "output_model", "output_tiledb"])
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/nextflow.config
new file mode 100644
index 0000000..c2f655b
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/integration/harmony_leiden'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Run harmony integration followed by neighbour calculations, leiden clustering and run umap on the result.'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/nextflow_schema.json
new file mode 100644
index 0000000..955281a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/nextflow_schema.json
@@ -0,0 +1,220 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "harmony_leiden",
+  "description": "Run harmony integration followed by neighbour calculations, leiden clustering and run umap on the result.",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to the sample",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"dataset.h5mu\"`. "
+        },
+        "tiledb_input_uri": {
+          "type": "string",
+          "description": "A URI containing the TileDB-SOMA objects",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"s3://bucket/path\"`. "
+        },
+        "tiledb_s3_region": {
+          "type": "string",
+          "description": "Region where the TileDB-SOMA database is hosted.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"us-west-2\"`. "
+        },
+        "tiledb_endpoint": {
+          "type": "string",
+          "description": "Custom endpoint to use to connect to S3\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "tiledb_s3_no_sign_request": {
+          "type": "boolean",
+          "description": "Do not sign S3 requests",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "output_tiledb": {
+          "type": "string",
+          "format": "path",
+          "description": "Output the TileDB database to the specified directory instead of adding it to the existing database.\n",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output_tiledb\"`, direction: `output`. ",
+          "default": "$id.$key.output_tiledb"
+        }
+      }
+    },
+    "input slots": {
+      "title": "Input slots",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "layer": {
+          "type": "string",
+          "description": "use specified layer for expression values instead of the .X object from the modality.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"log_normalized\"`. ",
+          "default": "log_normalized"
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality to process.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        }
+      }
+    },
+    "neighbour calculation": {
+      "title": "Neighbour calculation",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "uns_neighbors": {
+          "type": "string",
+          "description": "In which .uns slot to store various neighbor output objects.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"harmonypy_integration_neighbors\"`. ",
+          "default": "harmonypy_integration_neighbors"
+        },
+        "obsp_neighbor_distances": {
+          "type": "string",
+          "description": "In which .obsp slot to store the distance matrix between the resulting neighbors.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"harmonypy_integration_distances\"`. ",
+          "default": "harmonypy_integration_distances"
+        },
+        "obsp_neighbor_connectivities": {
+          "type": "string",
+          "description": "In which .obsp slot to store the connectivities matrix between the resulting neighbors.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"harmonypy_integration_connectivities\"`. ",
+          "default": "harmonypy_integration_connectivities"
+        }
+      }
+    },
+    "harmony integration options": {
+      "title": "Harmony integration options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "embedding": {
+          "type": "string",
+          "description": "Embedding to use as input",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_pca\"`. ",
+          "default": "X_pca"
+        },
+        "obsm_integrated": {
+          "type": "string",
+          "description": "In which .obsm slot to store the resulting integrated embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_pca_integrated\"`. ",
+          "default": "X_pca_integrated"
+        },
+        "obs_covariates": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The .obs field(s) that define the covariate(s) to regress out.",
+          "help_text": "Type: `string`, multiple: `True`, required, example: `[\"batch\";\"sample\"]`. "
+        },
+        "theta": {
+          "type": "array",
+          "items": {
+            "type": "number"
+          },
+          "description": "Diversity clustering penalty parameter",
+          "help_text": "Type: `double`, multiple: `True`, default: `[2.0]`. ",
+          "default": [
+            2.0
+          ]
+        }
+      }
+    },
+    "clustering options": {
+      "title": "Clustering options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "obs_cluster": {
+          "type": "string",
+          "description": "Prefix for the .obs keys under which to add the cluster labels",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"harmony_integration_leiden\"`. ",
+          "default": "harmony_integration_leiden"
+        },
+        "leiden_resolution": {
+          "type": "array",
+          "items": {
+            "type": "number"
+          },
+          "description": "Control the coarseness of the clustering",
+          "help_text": "Type: `double`, multiple: `True`, default: `[1.0]`. ",
+          "default": [
+            1.0
+          ]
+        }
+      }
+    },
+    "umap options": {
+      "title": "Umap options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "obsm_umap": {
+          "type": "string",
+          "description": "In which .obsm slot to store the resulting UMAP embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_leiden_harmony_umap\"`. ",
+          "default": "X_leiden_harmony_umap"
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/input slots"
+    },
+    {
+      "$ref": "#/$defs/neighbour calculation"
+    },
+    {
+      "$ref": "#/$defs/harmony integration options"
+    },
+    {
+      "$ref": "#/$defs/clustering options"
+    },
+    {
+      "$ref": "#/$defs/umap options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/.config.vsh.yaml
new file mode 100644
index 0000000..9fe296e
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/.config.vsh.yaml
@@ -0,0 +1,540 @@
+name: "scvi_leiden"
+namespace: "workflows/integration"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Path to the sample. Mutually exclusive with the 'input_uri' argument."
+    info: null
+    example:
+    - "dataset.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--tiledb_input_uri"
+    description: "A URI containing the TileDB-SOMA objects. Mutually exclusive with\
+      \ the 'input' argument."
+    info: null
+    example:
+    - "s3://bucket/path"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--tiledb_s3_region"
+    description: "Region where the TileDB-SOMA database is hosted.\n"
+    info: null
+    example:
+    - "us-west-2"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--tiledb_endpoint"
+    description: "Custom endpoint to use to connect to S3\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--tiledb_s3_no_sign_request"
+    description: "Do not sign S3 requests. Credentials will not be loaded if this\
+      \ argument is provided.\n"
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Input slots"
+  arguments:
+  - type: "string"
+    name: "--layer"
+    description: "use specified layer for expression values instead of the .X object\
+      \ from the modality."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--sanitize_ensembl_ids"
+    description: "Whether to sanitize ensembl ids by removing version numbers."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_model"
+    description: "Folder where the state of the trained model will be saved to."
+    info: null
+    example:
+    - "output_dir"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_tiledb"
+    description: "Output the TileDB database to the specified directory instead of\
+      \ adding it to the existing database.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Neighbour calculation"
+  arguments:
+  - type: "string"
+    name: "--uns_neighbors"
+    description: "In which .uns slot to store various neighbor output objects."
+    info: null
+    default:
+    - "scvi_integration_neighbors"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsp_neighbor_distances"
+    description: "In which .obsp slot to store the distance matrix between the resulting\
+      \ neighbors."
+    info: null
+    default:
+    - "scvi_integration_distances"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsp_neighbor_connectivities"
+    description: "In which .obsp slot to store the connectivities matrix between the\
+      \ resulting neighbors."
+    info: null
+    default:
+    - "scvi_integration_connectivities"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Scvi integration options"
+  arguments:
+  - type: "string"
+    name: "--obs_batch"
+    description: "Column name discriminating between your batches."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_size_factor"
+    description: "Key in adata.obs for size factor information. Instead of using library\
+      \ size as a size factor,\nthe provided size factor column will be used as offset\
+      \ in the mean of the likelihood.\nAssumed to be on linear scale.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_categorical_covariate"
+    description: "Keys in adata.obs that correspond to categorical data. These covariates\
+      \ can be added in\naddition to the batch covariate and are also treated as nuisance\
+      \ factors\n(i.e., the model tries to minimize their effects on the latent space).\n\
+      Thus, these should not be used for biologically-relevant factors that you do\
+      \ _not_ want to correct for.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_continuous_covariate"
+    description: "Keys in adata.obs that correspond to continuous data. These covariates\
+      \ can be added in\naddition to the batch covariate and are also treated as nuisance\
+      \ factors\n(i.e., the model tries to minimize their effects on the latent space).\
+      \ Thus, these should not be\nused for biologically-relevant factors that you\
+      \ do _not_ want to correct for.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_output"
+    description: "In which .obsm slot to store the resulting integrated embedding."
+    info: null
+    default:
+    - "X_scvi_integrated"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_input"
+    description: ".var column containing highly variable genes. By default, do not\
+      \ subset genes."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--early_stopping"
+    description: "Whether to perform early stopping with respect to the validation\
+      \ set."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--early_stopping_monitor"
+    description: "Metric logged during validation set epoch."
+    info: null
+    default:
+    - "elbo_validation"
+    required: false
+    choices:
+    - "elbo_validation"
+    - "reconstruction_loss_validation"
+    - "kl_local_validation"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--early_stopping_patience"
+    description: "Number of validation epochs with no improvement after which training\
+      \ will be stopped."
+    info: null
+    default:
+    - 45
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--early_stopping_min_delta"
+    description: "Minimum change in the monitored quantity to qualify as an improvement,\
+      \ i.e. an absolute change of less than min_delta, will count as no improvement."
+    info: null
+    default:
+    - 0.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_epochs"
+    description: "Number of passes through the dataset, defaults to (20000 / number\
+      \ of cells) * 400 or 400; whichever is smallest."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--reduce_lr_on_plateau"
+    description: "Whether to monitor validation loss and reduce learning rate when\
+      \ validation set `lr_scheduler_metric` plateaus."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--lr_factor"
+    description: "Factor to reduce learning rate."
+    info: null
+    default:
+    - 0.6
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--lr_patience"
+    description: "Number of epochs with no improvement after which learning rate will\
+      \ be reduced."
+    info: null
+    default:
+    - 30.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Clustering options"
+  arguments:
+  - type: "string"
+    name: "--obs_cluster"
+    description: "Prefix for the .obs keys under which to add the cluster labels.\
+      \ Newly created columns in .obs will \nbe created from the specified value for\
+      \ '--obs_cluster' suffixed with an underscore and one of the resolutions\nresolutions\
+      \ specified in '--leiden_resolution'.\n"
+    info: null
+    default:
+    - "scvi_integration_leiden"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--leiden_resolution"
+    description: "Control the coarseness of the clustering. Higher values lead to\
+      \ more clusters."
+    info: null
+    default:
+    - 1.0
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Umap options"
+  arguments:
+  - type: "string"
+    name: "--obsm_umap"
+    description: "In which .obsm slot to store the resulting UMAP embedding."
+    info: null
+    default:
+    - "X_scvi_umap"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Run scvi integration followed by neighbour calculations, leiden clustering\
+  \ and run umap on the result."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_tiledb_wf"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "integrate/scvi"
+  repository:
+    type: "local"
+- name: "workflows/multiomics/neighbors_leiden_umap"
+  repository:
+    type: "local"
+- name: "convert/from_tiledb_to_h5mu"
+  repository:
+    type: "local"
+- name: "tiledb/move_mudata_obsm_to_tiledb"
+  repository:
+    type: "local"
+- name: "tiledb/move_mudata_obs_to_tiledb"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/integration/scvi_leiden/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/integration/scvi_leiden"
+  executable: "target/nextflow/workflows/integration/scvi_leiden/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/integrate/scvi"
+  - "target/nextflow/workflows/multiomics/neighbors_leiden_umap"
+  - "target/nextflow/convert/from_tiledb_to_h5mu"
+  - "target/_private/nextflow/tiledb/move_mudata_obsm_to_tiledb"
+  - "target/_private/nextflow/tiledb/move_mudata_obs_to_tiledb"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/main.nf
new file mode 100644
index 0000000..a478b6b
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/main.nf
@@ -0,0 +1,3938 @@
+// scvi_leiden v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "scvi_leiden",
+  "namespace" : "workflows/integration",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Path to the sample. Mutually exclusive with the 'input_uri' argument.",
+          "example" : [
+            "dataset.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--tiledb_input_uri",
+          "description" : "A URI containing the TileDB-SOMA objects. Mutually exclusive with the 'input' argument.",
+          "example" : [
+            "s3://bucket/path"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--tiledb_s3_region",
+          "description" : "Region where the TileDB-SOMA database is hosted.\n",
+          "example" : [
+            "us-west-2"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--tiledb_endpoint",
+          "description" : "Custom endpoint to use to connect to S3\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--tiledb_s3_no_sign_request",
+          "description" : "Do not sign S3 requests. Credentials will not be loaded if this argument is provided.\n",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Input slots",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "use specified layer for expression values instead of the .X object from the modality.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--sanitize_ensembl_ids",
+          "description" : "Whether to sanitize ensembl ids by removing version numbers.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_model",
+          "description" : "Folder where the state of the trained model will be saved to.",
+          "example" : [
+            "output_dir"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_tiledb",
+          "description" : "Output the TileDB database to the specified directory instead of adding it to the existing database.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Neighbour calculation",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--uns_neighbors",
+          "description" : "In which .uns slot to store various neighbor output objects.",
+          "default" : [
+            "scvi_integration_neighbors"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsp_neighbor_distances",
+          "description" : "In which .obsp slot to store the distance matrix between the resulting neighbors.",
+          "default" : [
+            "scvi_integration_distances"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsp_neighbor_connectivities",
+          "description" : "In which .obsp slot to store the connectivities matrix between the resulting neighbors.",
+          "default" : [
+            "scvi_integration_connectivities"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Scvi integration options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obs_batch",
+          "description" : "Column name discriminating between your batches.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_size_factor",
+          "description" : "Key in adata.obs for size factor information. Instead of using library size as a size factor,\nthe provided size factor column will be used as offset in the mean of the likelihood.\nAssumed to be on linear scale.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_categorical_covariate",
+          "description" : "Keys in adata.obs that correspond to categorical data. These covariates can be added in\naddition to the batch covariate and are also treated as nuisance factors\n(i.e., the model tries to minimize their effects on the latent space).\nThus, these should not be used for biologically-relevant factors that you do _not_ want to correct for.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_continuous_covariate",
+          "description" : "Keys in adata.obs that correspond to continuous data. These covariates can be added in\naddition to the batch covariate and are also treated as nuisance factors\n(i.e., the model tries to minimize their effects on the latent space). Thus, these should not be\nused for biologically-relevant factors that you do _not_ want to correct for.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_output",
+          "description" : "In which .obsm slot to store the resulting integrated embedding.",
+          "default" : [
+            "X_scvi_integrated"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_input",
+          "description" : ".var column containing highly variable genes. By default, do not subset genes.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--early_stopping",
+          "description" : "Whether to perform early stopping with respect to the validation set.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--early_stopping_monitor",
+          "description" : "Metric logged during validation set epoch.",
+          "default" : [
+            "elbo_validation"
+          ],
+          "required" : false,
+          "choices" : [
+            "elbo_validation",
+            "reconstruction_loss_validation",
+            "kl_local_validation"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--early_stopping_patience",
+          "description" : "Number of validation epochs with no improvement after which training will be stopped.",
+          "default" : [
+            45
+          ],
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--early_stopping_min_delta",
+          "description" : "Minimum change in the monitored quantity to qualify as an improvement, i.e. an absolute change of less than min_delta, will count as no improvement.",
+          "default" : [
+            0.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_epochs",
+          "description" : "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--reduce_lr_on_plateau",
+          "description" : "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--lr_factor",
+          "description" : "Factor to reduce learning rate.",
+          "default" : [
+            0.6
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--lr_patience",
+          "description" : "Number of epochs with no improvement after which learning rate will be reduced.",
+          "default" : [
+            30.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Clustering options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obs_cluster",
+          "description" : "Prefix for the .obs keys under which to add the cluster labels. Newly created columns in .obs will \nbe created from the specified value for '--obs_cluster' suffixed with an underscore and one of the resolutions\nresolutions specified in '--leiden_resolution'.\n",
+          "default" : [
+            "scvi_integration_leiden"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--leiden_resolution",
+          "description" : "Control the coarseness of the clustering. Higher values lead to more clusters.",
+          "default" : [
+            1.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Umap options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obsm_umap",
+          "description" : "In which .obsm slot to store the resulting UMAP embedding.",
+          "default" : [
+            "X_scvi_umap"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Run scvi integration followed by neighbour calculations, leiden clustering and run umap on the result.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_tiledb_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "integrate/scvi",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/multiomics/neighbors_leiden_umap",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "convert/from_tiledb_to_h5mu",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "tiledb/move_mudata_obsm_to_tiledb",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "tiledb/move_mudata_obs_to_tiledb",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/integration/scvi_leiden/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/integration/scvi_leiden",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { scvi } from "${meta.resources_dir}/../../../../nextflow/integrate/scvi/main.nf"
+include { neighbors_leiden_umap } from "${meta.resources_dir}/../../../../nextflow/workflows/multiomics/neighbors_leiden_umap/main.nf"
+include { from_tiledb_to_h5mu } from "${meta.resources_dir}/../../../../nextflow/convert/from_tiledb_to_h5mu/main.nf"
+include { move_mudata_obsm_to_tiledb } from "${meta.resources_dir}/../../../../_private/nextflow/tiledb/move_mudata_obsm_to_tiledb/main.nf"
+include { move_mudata_obs_to_tiledb } from "${meta.resources_dir}/../../../../_private/nextflow/tiledb/move_mudata_obs_to_tiledb/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+  output_ch = input_ch
+    | map {id, state -> 
+      def new_state = state + [
+        "workflow_output": state.output,
+        "output_tiledb": state.tiledb_input_uri ? state.output_tiledb : null
+      ]
+      [id, new_state]
+    }
+    | map {id, state -> 
+      if (!workflow.stubRun) {
+        assert state.input || state.tiledb_input_uri: "Either --input or --tiledb_input_uri must be defined"
+        assert !(state.input && state.tiledb_input_uri): "Values were specified for both --input and --tiledb_input_uri. Please choose one."
+        assert state.tiledb_s3_region || !state.tiledb_input_uri: "Specifying 'tiledb_s3_region' also requires 'tiledb_input_uri'."
+        assert !state.tiledb_input_uri || state.layer: "When using tileDB input, you must specify a layer using --layer"
+      }
+      [id, state]
+    }
+    | from_tiledb_to_h5mu.run(
+      runIf: {id, state -> state.tiledb_input_uri},
+      fromState: { id, state -> 
+        [
+          "input_uri": state.tiledb_input_uri,
+          "s3_region": state.tiledb_s3_region,
+          "endpoint": state.tiledb_endpoint,
+          "input_modality": state.modality,
+          "output_modality": state.modality,
+          "input_layers": [ state.layer ],
+          "s3_no_sign_request": state.tiledb_s3_no_sign_request
+        ]
+      },
+      toState: [
+        "input": "output",
+      ]
+    )
+    | scvi.run(
+      fromState: [
+        "input": "input",
+        "obs_batch": "obs_batch",
+        "obs_size_factor": "obs_size_factor",
+        "obs_categorical_covariate": "obs_categorical_covariate",
+        "obs_continuous_covariate": "obs_continuous_covariate",
+        "obsm_output": "obsm_output",
+        "var_input": "var_input",
+        "early_stopping": "early_stopping",
+        "early_stopping_monitor": "early_stopping_monitor",
+        "early_stopping_patience": "early_stopping_patience",
+        "early_stopping_min_delta": "early_stopping_min_delta",
+        "max_epochs": "max_epochs",
+        "reduce_lr_on_plateau": "reduce_lr_on_plateau",
+        "lr_factor": "lr_factor",
+        "lr_patience": "lr_patience",
+        "output_model": "output_model",
+        "modality": "modality",
+        "input_layer": "layer",
+        "sanitize_ensembl_ids": "sanitize_ensembl_ids"
+      ],
+      toState: [
+        "input": "output", 
+        "output_model": "output_model"
+      ],
+    )
+    | neighbors_leiden_umap.run(
+      fromState: [
+        "input": "input",
+        "modality": "modality",
+        "obsm_input": "obsm_output",
+        "output": "workflow_output",
+        "uns_neighbors": "uns_neighbors",
+        "obsp_neighbor_distances": "obsp_neighbor_distances",
+        "obsp_neighbor_connectivities": "obsp_neighbor_connectivities",
+        "leiden_resolution": "leiden_resolution",
+        "obs_cluster": "obs_cluster",
+        "obsm_umap": "obsm_umap",
+      ],
+      toState: ["output": "output"]
+    )
+    | move_mudata_obsm_to_tiledb.run(
+      runIf: {id, state -> state.tiledb_input_uri},
+      fromState: {id, state ->
+        [
+          "input_mudata": state.output,
+          "modality": state.modality,
+          "input_uri": state.tiledb_input_uri,
+          "s3_region": state.tiledb_s3_region,
+          "endpoint": state.tiledb_endpoint,
+          "output_modality": state.modality,
+          "obsm_input": [state.obsm_umap, state.obsm_output],
+          "output_tiledb": state.output_tiledb,
+          "s3_no_sign_request": state.tiledb_s3_no_sign_request
+        ]
+      },
+      toState: ["previous_output_tiledb": "output_tiledb"]
+    )
+    | move_mudata_obs_to_tiledb.run(
+      runIf: {id, state -> state.tiledb_input_uri},
+      fromState: {id, state ->
+        def new_state = [
+          "input_mudata": state.output,
+          "modality": state.modality,
+          "s3_region": state.tiledb_s3_region,
+          "endpoint": state.tiledb_endpoint,
+          "output_modality": state.modality,
+          "obs_input": state.leiden_resolution.collect{state.obs_cluster + "_" + it},
+          "output_tiledb": state.output_tiledb,
+          "s3_no_sign_request": state.tiledb_s3_no_sign_request
+        ]
+        if (state.previous_output_tiledb && file(state.previous_output_tiledb).exists()) {
+          new_state += ["input_dir": state.previous_output_tiledb]
+        } else {
+          new_state += ["input_uri": state.tiledb_input_uri]
+        }
+        return new_state
+      },
+      toState: ["output_tiledb": "output_tiledb"]
+    )
+    | setState(["output", "output_model", "output_tiledb"])
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/nextflow.config
new file mode 100644
index 0000000..e4b6f8b
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/integration/scvi_leiden'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Run scvi integration followed by neighbour calculations, leiden clustering and run umap on the result.'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/nextflow_schema.json
new file mode 100644
index 0000000..4f3b70a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/nextflow_schema.json
@@ -0,0 +1,289 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "scvi_leiden",
+  "description": "Run scvi integration followed by neighbour calculations, leiden clustering and run umap on the result.",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to the sample",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"dataset.h5mu\"`. "
+        },
+        "tiledb_input_uri": {
+          "type": "string",
+          "description": "A URI containing the TileDB-SOMA objects",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"s3://bucket/path\"`. "
+        },
+        "tiledb_s3_region": {
+          "type": "string",
+          "description": "Region where the TileDB-SOMA database is hosted.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"us-west-2\"`. "
+        },
+        "tiledb_endpoint": {
+          "type": "string",
+          "description": "Custom endpoint to use to connect to S3\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "tiledb_s3_no_sign_request": {
+          "type": "boolean",
+          "description": "Do not sign S3 requests",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        },
+        "output_model": {
+          "type": "string",
+          "format": "path",
+          "description": "Folder where the state of the trained model will be saved to.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output_model\"`, direction: `output`, example: `\"output_dir\"`. ",
+          "default": "$id.$key.output_model"
+        },
+        "output_tiledb": {
+          "type": "string",
+          "format": "path",
+          "description": "Output the TileDB database to the specified directory instead of adding it to the existing database.\n",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output_tiledb\"`, direction: `output`. ",
+          "default": "$id.$key.output_tiledb"
+        }
+      }
+    },
+    "input slots": {
+      "title": "Input slots",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "layer": {
+          "type": "string",
+          "description": "use specified layer for expression values instead of the .X object from the modality.",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality to process.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "sanitize_ensembl_ids": {
+          "type": "boolean",
+          "description": "Whether to sanitize ensembl ids by removing version numbers.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "neighbour calculation": {
+      "title": "Neighbour calculation",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "uns_neighbors": {
+          "type": "string",
+          "description": "In which .uns slot to store various neighbor output objects.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scvi_integration_neighbors\"`. ",
+          "default": "scvi_integration_neighbors"
+        },
+        "obsp_neighbor_distances": {
+          "type": "string",
+          "description": "In which .obsp slot to store the distance matrix between the resulting neighbors.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scvi_integration_distances\"`. ",
+          "default": "scvi_integration_distances"
+        },
+        "obsp_neighbor_connectivities": {
+          "type": "string",
+          "description": "In which .obsp slot to store the connectivities matrix between the resulting neighbors.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scvi_integration_connectivities\"`. ",
+          "default": "scvi_integration_connectivities"
+        }
+      }
+    },
+    "scvi integration options": {
+      "title": "Scvi integration options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "obs_batch": {
+          "type": "string",
+          "description": "Column name discriminating between your batches.",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        },
+        "obs_size_factor": {
+          "type": "string",
+          "description": "Key in adata.obs for size factor information",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_categorical_covariate": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys in adata.obs that correspond to categorical data",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "obs_continuous_covariate": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys in adata.obs that correspond to continuous data",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "obsm_output": {
+          "type": "string",
+          "description": "In which .obsm slot to store the resulting integrated embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_scvi_integrated\"`. ",
+          "default": "X_scvi_integrated"
+        },
+        "var_input": {
+          "type": "string",
+          "description": ".var column containing highly variable genes",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "early_stopping": {
+          "type": "boolean",
+          "description": "Whether to perform early stopping with respect to the validation set.",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        },
+        "early_stopping_monitor": {
+          "type": "string",
+          "description": "Metric logged during validation set epoch.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"elbo_validation\"`, choices: ``elbo_validation`, `reconstruction_loss_validation`, `kl_local_validation``. ",
+          "enum": [
+            "elbo_validation",
+            "reconstruction_loss_validation",
+            "kl_local_validation"
+          ],
+          "default": "elbo_validation"
+        },
+        "early_stopping_patience": {
+          "type": "integer",
+          "description": "Number of validation epochs with no improvement after which training will be stopped.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `45`. ",
+          "default": 45
+        },
+        "early_stopping_min_delta": {
+          "type": "number",
+          "description": "Minimum change in the monitored quantity to qualify as an improvement, i.e",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.0`. ",
+          "default": 0.0
+        },
+        "max_epochs": {
+          "type": "integer",
+          "description": "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest.",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "reduce_lr_on_plateau": {
+          "type": "boolean",
+          "description": "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "lr_factor": {
+          "type": "number",
+          "description": "Factor to reduce learning rate.",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.6`. ",
+          "default": 0.6
+        },
+        "lr_patience": {
+          "type": "number",
+          "description": "Number of epochs with no improvement after which learning rate will be reduced.",
+          "help_text": "Type: `double`, multiple: `False`, default: `30.0`. ",
+          "default": 30.0
+        }
+      }
+    },
+    "clustering options": {
+      "title": "Clustering options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "obs_cluster": {
+          "type": "string",
+          "description": "Prefix for the .obs keys under which to add the cluster labels",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scvi_integration_leiden\"`. ",
+          "default": "scvi_integration_leiden"
+        },
+        "leiden_resolution": {
+          "type": "array",
+          "items": {
+            "type": "number"
+          },
+          "description": "Control the coarseness of the clustering",
+          "help_text": "Type: `double`, multiple: `True`, default: `[1.0]`. ",
+          "default": [
+            1.0
+          ]
+        }
+      }
+    },
+    "umap options": {
+      "title": "Umap options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "obsm_umap": {
+          "type": "string",
+          "description": "In which .obsm slot to store the resulting UMAP embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_scvi_umap\"`. ",
+          "default": "X_scvi_umap"
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/input slots"
+    },
+    {
+      "$ref": "#/$defs/neighbour calculation"
+    },
+    {
+      "$ref": "#/$defs/scvi integration options"
+    },
+    {
+      "$ref": "#/$defs/clustering options"
+    },
+    {
+      "$ref": "#/$defs/umap options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/.config.vsh.yaml
new file mode 100644
index 0000000..782dac2
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/.config.vsh.yaml
@@ -0,0 +1,329 @@
+name: "dimensionality_reduction"
+namespace: "workflows/multiomics"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the sample."
+    info: null
+    example:
+    - "foo"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    description: "Path to the sample."
+    info: null
+    example:
+    - "dataset.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "use specified layer for expression values instead of the .X object\
+      \ from the modality."
+    info: null
+    default:
+    - "log_normalized"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "PCA options"
+  arguments:
+  - type: "string"
+    name: "--obsm_pca"
+    description: "In which .obsm slot to store the resulting PCA embedding."
+    info: null
+    default:
+    - "X_pca"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_pca_feature_selection"
+    description: "Column name in .var matrix that will be used to select which genes\
+      \ to run the PCA on."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--pca_loadings_varm_output"
+    description: "Name of the .varm key where the PCA loadings are stored.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--pca_variance_uns_output"
+    description: "Name of the .uns key where the variance and variance ratio will\
+      \ be stored as a map.\nThe map will contain two keys: variance and variance_ratio\
+      \ respectively.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--pca_overwrite"
+    description: "Allow overwriting slots for PCA output."
+    info: null
+    direction: "input"
+- name: "Neighbour calculation"
+  arguments:
+  - type: "string"
+    name: "--uns_neighbors"
+    description: "In which .uns slot to store various neighbor output objects."
+    info: null
+    default:
+    - "neighbors"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsp_neighbor_distances"
+    description: "In which .obsp slot to store the distance matrix between the resulting\
+      \ neighbors."
+    info: null
+    default:
+    - "distances"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsp_neighbor_connectivities"
+    description: "In which .obsp slot to store the connectivities matrix between the\
+      \ resulting neighbors."
+    info: null
+    default:
+    - "connectivities"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Umap options"
+  arguments:
+  - type: "string"
+    name: "--obsm_umap"
+    description: "In which .obsm slot to store the resulting UMAP embedding."
+    info: null
+    default:
+    - "X_umap"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Run calculations that output information required for most integration\
+  \ methods: PCA, nearest neighbour and UMAP."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "concat_test_data"
+info:
+  test_dependencies:
+  - name: "dimensionality_reduction_test"
+    namespace: "test_workflows/multiomics"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "dimred/pca"
+  repository:
+    type: "local"
+- name: "workflows/multiomics/neighbors_leiden_umap"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/multiomics/dimensionality_reduction/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/multiomics/dimensionality_reduction"
+  executable: "target/nextflow/workflows/multiomics/dimensionality_reduction/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/dimred/pca"
+  - "target/nextflow/workflows/multiomics/neighbors_leiden_umap"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/main.nf
new file mode 100644
index 0000000..0dcf0e7
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/main.nf
@@ -0,0 +1,3626 @@
+// dimensionality_reduction v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "dimensionality_reduction",
+  "namespace" : "workflows/multiomics",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the sample.",
+          "example" : [
+            "foo"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Path to the sample.",
+          "example" : [
+            "dataset.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "use specified layer for expression values instead of the .X object from the modality.",
+          "default" : [
+            "log_normalized"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "PCA options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obsm_pca",
+          "description" : "In which .obsm slot to store the resulting PCA embedding.",
+          "default" : [
+            "X_pca"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_pca_feature_selection",
+          "description" : "Column name in .var matrix that will be used to select which genes to run the PCA on.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--pca_loadings_varm_output",
+          "description" : "Name of the .varm key where the PCA loadings are stored.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--pca_variance_uns_output",
+          "description" : "Name of the .uns key where the variance and variance ratio will be stored as a map.\nThe map will contain two keys: variance and variance_ratio respectively.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--pca_overwrite",
+          "description" : "Allow overwriting slots for PCA output.",
+          "direction" : "input"
+        }
+      ]
+    },
+    {
+      "name" : "Neighbour calculation",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--uns_neighbors",
+          "description" : "In which .uns slot to store various neighbor output objects.",
+          "default" : [
+            "neighbors"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsp_neighbor_distances",
+          "description" : "In which .obsp slot to store the distance matrix between the resulting neighbors.",
+          "default" : [
+            "distances"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsp_neighbor_connectivities",
+          "description" : "In which .obsp slot to store the connectivities matrix between the resulting neighbors.",
+          "default" : [
+            "connectivities"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Umap options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obsm_umap",
+          "description" : "In which .obsm slot to store the resulting UMAP embedding.",
+          "default" : [
+            "X_umap"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Run calculations that output information required for most integration methods: PCA, nearest neighbour and UMAP.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data"
+    }
+  ],
+  "info" : {
+    "test_dependencies" : [
+      {
+        "name" : "dimensionality_reduction_test",
+        "namespace" : "test_workflows/multiomics"
+      }
+    ]
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "dimred/pca",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/multiomics/neighbors_leiden_umap",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/multiomics/dimensionality_reduction/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/multiomics/dimensionality_reduction",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { pca } from "${meta.resources_dir}/../../../../nextflow/dimred/pca/main.nf"
+include { neighbors_leiden_umap } from "${meta.resources_dir}/../../../../nextflow/workflows/multiomics/neighbors_leiden_umap/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+  output_ch = input_ch
+    | map {id, state -> 
+      def new_state = state + ["workflow_output": state.output]
+      [id, new_state]
+    }
+    | pca.run(
+      fromState: [
+        "input": "input", 
+        "obsm_output": "obsm_pca",
+        "var_input": "var_pca_feature_selection",
+        "modality": "modality",
+        "overwrite": "pca_overwrite",
+        "layer": "layer",
+        "varm_output": "pca_loadings_varm_output",
+        "uns_output": "pca_variance_uns_output",
+      ],
+      toState: ["input": "output"]
+    )
+    | neighbors_leiden_umap.run(
+      fromState: [
+        "input": "input",
+        "obsm_input": "obsm_pca",
+        "modality": "modality",
+        "uns_neighbors": "uns_neighbors",
+        "obsp_neighbor_distances": "obsp_neighbor_distances",
+        "obsp_neighbor_connectivities": "obsp_neighbor_connectivities",
+        "output": "workflow_output",
+        "obsm_umap": "obsm_umap",
+      ],
+      toState: ["output": "output"],
+      args: [
+        "leiden_resolution": [] // disable leiden
+      ]
+    )
+    | setState(["output"])
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/nextflow.config
new file mode 100644
index 0000000..03ff143
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/multiomics/dimensionality_reduction'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Run calculations that output information required for most integration methods: PCA, nearest neighbour and UMAP.'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/nextflow_schema.json
new file mode 100644
index 0000000..d241601
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/nextflow_schema.json
@@ -0,0 +1,157 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "dimensionality_reduction",
+  "description": "Run calculations that output information required for most integration methods: PCA, nearest neighbour and UMAP.",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "ID of the sample.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"foo\"`. "
+        },
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the sample.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"dataset.h5mu\"`. "
+        },
+        "layer": {
+          "type": "string",
+          "description": "use specified layer for expression values instead of the .X object from the modality.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"log_normalized\"`. ",
+          "default": "log_normalized"
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality to process.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "pca options": {
+      "title": "PCA options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "obsm_pca": {
+          "type": "string",
+          "description": "In which .obsm slot to store the resulting PCA embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_pca\"`. ",
+          "default": "X_pca"
+        },
+        "var_pca_feature_selection": {
+          "type": "string",
+          "description": "Column name in .var matrix that will be used to select which genes to run the PCA on.",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "pca_loadings_varm_output": {
+          "type": "string",
+          "description": "Name of the .varm key where the PCA loadings are stored.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "pca_variance_uns_output": {
+          "type": "string",
+          "description": "Name of the .uns key where the variance and variance ratio will be stored as a map.\nThe map will contain two keys: variance and variance_ratio respectively.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "pca_overwrite": {
+          "type": "boolean",
+          "description": "Allow overwriting slots for PCA output.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "neighbour calculation": {
+      "title": "Neighbour calculation",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "uns_neighbors": {
+          "type": "string",
+          "description": "In which .uns slot to store various neighbor output objects.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"neighbors\"`. ",
+          "default": "neighbors"
+        },
+        "obsp_neighbor_distances": {
+          "type": "string",
+          "description": "In which .obsp slot to store the distance matrix between the resulting neighbors.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"distances\"`. ",
+          "default": "distances"
+        },
+        "obsp_neighbor_connectivities": {
+          "type": "string",
+          "description": "In which .obsp slot to store the connectivities matrix between the resulting neighbors.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"connectivities\"`. ",
+          "default": "connectivities"
+        }
+      }
+    },
+    "umap options": {
+      "title": "Umap options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "obsm_umap": {
+          "type": "string",
+          "description": "In which .obsm slot to store the resulting UMAP embedding.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"X_umap\"`. ",
+          "default": "X_umap"
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/pca options"
+    },
+    {
+      "$ref": "#/$defs/neighbour calculation"
+    },
+    {
+      "$ref": "#/$defs/umap options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/dimensionality_reduction/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/.config.vsh.yaml
new file mode 100644
index 0000000..3991df7
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/.config.vsh.yaml
@@ -0,0 +1,287 @@
+name: "neighbors_leiden_umap"
+namespace: "workflows/multiomics"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "Path to the sample."
+    info: null
+    example:
+    - "dataset.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsm_input"
+    description: "The key of the embedding to use as input."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Neighbour calculation"
+  arguments:
+  - type: "string"
+    name: "--uns_neighbors"
+    description: "In which .uns slot to store various neighbor output objects."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsp_neighbor_distances"
+    description: "In which .obsp slot to store the distance matrix between the resulting\
+      \ neighbors."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obsp_neighbor_connectivities"
+    description: "In which .obsp slot to store the connectivities matrix between the\
+      \ resulting neighbors."
+    info: null
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Clustering options"
+  arguments:
+  - type: "string"
+    name: "--obs_cluster"
+    description: "Prefix for the .obs keys under which to add the cluster labels.\
+      \ Newly created columns in .obs will \nbe created from the specified value for\
+      \ '--obs_cluster' suffixed with an underscore and one of the resolutions\nresolutions\
+      \ specified in '--leiden_resolution'.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--leiden_resolution"
+    description: "Control the coarseness of the clustering. Higher values lead to\
+      \ more clusters."
+    info: null
+    default:
+    - 1.0
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Umap options"
+  arguments:
+  - type: "string"
+    name: "--obsm_umap"
+    description: "In which .obsm slot to store the resulting UMAP embedding.\nWhen\
+      \ not specified, UMAP will not be executed.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Performs neighborhood search, leiden clustering and run umap on an integrated\
+  \ embedding."
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "cluster/leiden"
+  repository:
+    type: "local"
+- name: "dimred/umap"
+  repository:
+    type: "local"
+- name: "neighbors/find_neighbors"
+  repository:
+    type: "local"
+- name: "metadata/move_obsm_to_obs"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/multiomics/neighbors_leiden_umap/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/multiomics/neighbors_leiden_umap"
+  executable: "target/nextflow/workflows/multiomics/neighbors_leiden_umap/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/cluster/leiden"
+  - "target/nextflow/dimred/umap"
+  - "target/nextflow/neighbors/find_neighbors"
+  - "target/nextflow/metadata/move_obsm_to_obs"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/main.nf
new file mode 100644
index 0000000..cd0c2da
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/main.nf
@@ -0,0 +1,3593 @@
+// neighbors_leiden_umap v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "neighbors_leiden_umap",
+  "namespace" : "workflows/multiomics",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Path to the sample.",
+          "example" : [
+            "dataset.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsm_input",
+          "description" : "The key of the embedding to use as input.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Neighbour calculation",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--uns_neighbors",
+          "description" : "In which .uns slot to store various neighbor output objects.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsp_neighbor_distances",
+          "description" : "In which .obsp slot to store the distance matrix between the resulting neighbors.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obsp_neighbor_connectivities",
+          "description" : "In which .obsp slot to store the connectivities matrix between the resulting neighbors.",
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Clustering options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obs_cluster",
+          "description" : "Prefix for the .obs keys under which to add the cluster labels. Newly created columns in .obs will \nbe created from the specified value for '--obs_cluster' suffixed with an underscore and one of the resolutions\nresolutions specified in '--leiden_resolution'.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--leiden_resolution",
+          "description" : "Control the coarseness of the clustering. Higher values lead to more clusters.",
+          "default" : [
+            1.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Umap options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obsm_umap",
+          "description" : "In which .obsm slot to store the resulting UMAP embedding.\nWhen not specified, UMAP will not be executed.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Performs neighborhood search, leiden clustering and run umap on an integrated embedding.",
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "cluster/leiden",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "dimred/umap",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "neighbors/find_neighbors",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "metadata/move_obsm_to_obs",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/multiomics/neighbors_leiden_umap/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/multiomics/neighbors_leiden_umap",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { leiden } from "${meta.resources_dir}/../../../../nextflow/cluster/leiden/main.nf"
+include { umap } from "${meta.resources_dir}/../../../../nextflow/dimred/umap/main.nf"
+include { find_neighbors } from "${meta.resources_dir}/../../../../nextflow/neighbors/find_neighbors/main.nf"
+include { move_obsm_to_obs } from "${meta.resources_dir}/../../../../nextflow/metadata/move_obsm_to_obs/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+  output_ch = input_ch
+    | map {id, state ->
+      if (!workflow.stubRun) {
+        assert (state.leiden_resolution.isEmpty() || state.obs_cluster?.trim()):
+          "When leiden_resolution is set, obs_cluster must also be defined."
+      }
+      [id, state]
+    }
+    | map {id, state ->
+      def new_state = state + ["workflow_output": state.output]
+      [id, new_state]
+    }
+    | find_neighbors.run(
+      fromState: [
+        "input": "input",
+        "uns_output": "uns_neighbors",
+        "obsp_distances": "obsp_neighbor_distances",
+        "obsp_connectivities": "obsp_neighbor_connectivities",
+        "obsm_input": "obsm_input",
+        "output": "workflow_output",
+        "modality": "modality"
+      ],
+      toState: ["input": "output"]
+    )
+    | leiden.run(
+      runIf: {id, state -> state.leiden_resolution},
+      fromState: [
+        "input": "input",
+        "obsp_connectivities": "obsp_neighbor_connectivities",
+        "obsm_name": "obs_cluster",
+        "resolution": "leiden_resolution",
+        "modality": "modality",
+      ],
+      toState: ["input": "output"]
+    )
+    | move_obsm_to_obs.run(
+      runIf: {id, state -> state.leiden_resolution},
+      fromState: [
+        "input": "input",
+        "output": "workflow_output",
+        "obsm_key": "obs_cluster",
+        "modality": "modality",
+      ],
+      args: ["output_compression": "gzip"],
+      toState: ["input": "output"]
+    )
+    | umap.run(
+      runIf: {id, state -> !state.obsm_umap?.trim()?.isEmpty()},
+      fromState: [
+          "input": "input",
+          "output": "workflow_output",
+          "uns_neighbors": "uns_neighbors",
+          "obsm_output": "obsm_umap",
+          "modality": "modality",
+        ],
+      args: ["output_compression": "gzip"],
+      toState: ["input": "output"]
+    )
+    | setState(["output": "input"])
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/nextflow.config
new file mode 100644
index 0000000..3c3148c
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/multiomics/neighbors_leiden_umap'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Performs neighborhood search, leiden clustering and run umap on an integrated embedding.'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/nextflow_schema.json
new file mode 100644
index 0000000..5c8c869
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/nextflow_schema.json
@@ -0,0 +1,136 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "neighbors_leiden_umap",
+  "description": "Performs neighborhood search, leiden clustering and run umap on an integrated embedding.",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the sample.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"dataset.h5mu\"`. "
+        },
+        "obsm_input": {
+          "type": "string",
+          "description": "The key of the embedding to use as input.",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality to process.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "neighbour calculation": {
+      "title": "Neighbour calculation",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "uns_neighbors": {
+          "type": "string",
+          "description": "In which .uns slot to store various neighbor output objects.",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        },
+        "obsp_neighbor_distances": {
+          "type": "string",
+          "description": "In which .obsp slot to store the distance matrix between the resulting neighbors.",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        },
+        "obsp_neighbor_connectivities": {
+          "type": "string",
+          "description": "In which .obsp slot to store the connectivities matrix between the resulting neighbors.",
+          "help_text": "Type: `string`, multiple: `False`, required. "
+        }
+      }
+    },
+    "clustering options": {
+      "title": "Clustering options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "obs_cluster": {
+          "type": "string",
+          "description": "Prefix for the .obs keys under which to add the cluster labels",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "leiden_resolution": {
+          "type": "array",
+          "items": {
+            "type": "number"
+          },
+          "description": "Control the coarseness of the clustering",
+          "help_text": "Type: `double`, multiple: `True`, default: `[1.0]`. ",
+          "default": [
+            1.0
+          ]
+        }
+      }
+    },
+    "umap options": {
+      "title": "Umap options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "obsm_umap": {
+          "type": "string",
+          "description": "In which .obsm slot to store the resulting UMAP embedding.\nWhen not specified, UMAP will not be executed.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/neighbour calculation"
+    },
+    {
+      "$ref": "#/$defs/clustering options"
+    },
+    {
+      "$ref": "#/$defs/umap options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/neighbors_leiden_umap/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/.config.vsh.yaml
new file mode 100644
index 0000000..05ee202
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/.config.vsh.yaml
@@ -0,0 +1,436 @@
+name: "process_batches"
+namespace: "workflows/multiomics"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the sample."
+    info: null
+    example:
+    - "foo"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to the sample."
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--rna_layer"
+    description: "Input layer for the gene expression modality. If not specified,\
+      \ .X is used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--prot_layer"
+    description: "Input layer for the antibody capture modality. If not specified,\
+      \ .X is used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Highly variable features detection"
+  arguments:
+  - type: "string"
+    name: "--highly_variable_features_var_output"
+    alternatives:
+    - "--filter_with_hvg_var_output"
+    description: "In which .var slot to store a boolean array corresponding to the\
+      \ highly variable genes."
+    info: null
+    default:
+    - "filter_with_hvg"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--highly_variable_features_obs_batch_key"
+    alternatives:
+    - "--filter_with_hvg_obs_batch_key"
+    description: "If specified, highly-variable genes are selected within each batch\
+      \ separately and merged. This simple \nprocess avoids the selection of batch-specific\
+      \ genes and acts as a lightweight batch correction method.\n"
+    info: null
+    default:
+    - "sample_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "QC metrics calculation options"
+  arguments:
+  - type: "string"
+    name: "--var_qc_metrics"
+    description: "Keys to select a boolean (containing only True or False) column\
+      \ from .var.\nFor each cell, calculate the proportion of total values for genes\
+      \ which are labeled 'True', \ncompared to the total sum of the values for all\
+      \ genes.\n"
+    info: null
+    example:
+    - "ercc,highly_variable"
+    default:
+    - "filter_with_hvg"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+  - type: "integer"
+    name: "--top_n_vars"
+    description: "Number of top vars to be used to calculate cumulative proportions.\n\
+      If not specified, proportions are not calculated. `--top_n_vars 20,50` finds\n\
+      cumulative proportion to the 20th and 50th most expressed vars.\n"
+    info: null
+    default:
+    - 50
+    - 100
+    - 200
+    - 500
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+- name: "PCA options"
+  arguments:
+  - type: "boolean_true"
+    name: "--pca_overwrite"
+    description: "Allow overwriting slots for PCA output."
+    info: null
+    direction: "input"
+- name: "CLR options"
+  arguments:
+  - type: "integer"
+    name: "--clr_axis"
+    description: "Axis to perform the CLR transformation on."
+    info: null
+    default:
+    - 0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "RNA Scaling options"
+  description: "Options for enabling scaling of the log-normalized data to unit variance\
+    \ and zero mean.\nThe scaled data will be output a different layer and representation\
+    \ with reduced dimensions\nwill be created and stored in addition to the non-scaled\
+    \ data.\n"
+  arguments:
+  - type: "boolean_true"
+    name: "--rna_enable_scaling"
+    description: "Enable scaling for the RNA modality."
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--rna_scaling_output_layer"
+    description: "Output layer where the scaled log-normalized data will be stored."
+    info: null
+    default:
+    - "scaled"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--rna_scaling_pca_obsm_output"
+    description: "Name of the .obsm key where the PCA representation of the log-normalized\n\
+      and scaled data is stored.\n"
+    info: null
+    default:
+    - "scaled_pca"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--rna_scaling_pca_loadings_varm_output"
+    description: "Name of the .varm key where the PCA loadings of the log-normalized\
+      \ and scaled\ndata is stored.\n"
+    info: null
+    default:
+    - "scaled_pca_loadings"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--rna_scaling_pca_variance_uns_output"
+    description: "Name of the .uns key where the variance and variance ratio will\
+      \ be stored as a map.\nThe map will contain two keys: variance and variance_ratio\
+      \ respectively.\n"
+    info: null
+    default:
+    - "scaled_pca_variance"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--rna_scaling_umap_obsm_output"
+    description: "Name of the .obsm key where the UMAP representation of the log-normalized\
+      \ and scaled data is stored."
+    info: null
+    default:
+    - "scaled_umap"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--rna_scaling_max_value"
+    description: "Clip (truncate) data to this value after scaling. If not specified,\
+      \ do not clip."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_false"
+    name: "--rna_scaling_zero_center"
+    description: "If set, omit zero-centering variables, which allows to handle sparse\
+      \ input efficiently.\""
+    info: null
+    direction: "input"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "This workflow serves as an entrypoint into the 'full_pipeline' in order\
+  \ to\nre-run the multisample processing and the integration setup. An input .h5mu\
+  \ file will \nfirst be split in order to run the multisample processing per modality.\
+  \ Next, the modalities\nare merged again and the integration setup pipeline is executed.\
+  \ Please note that this workflow\nassumes that samples from multiple pipelines are\
+  \ already concatenated. \n"
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf2"
+- type: "file"
+  path: "concat_test_data"
+- type: "file"
+  path: "10x_5k_anticmv"
+info:
+  test_dependencies:
+  - name: "workflow_test"
+    namespace: "test_workflows/multiomics/process_batches"
+  - name: "workflow_test2"
+    namespace: "test_workflows/multiomics/process_batches"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "dataflow/merge"
+  repository:
+    type: "local"
+- name: "workflows/multiomics/split_modalities"
+  alias: "split_modalities_workflow"
+  repository:
+    type: "local"
+- name: "workflows/prot/prot_multisample"
+  repository:
+    type: "local"
+- name: "workflows/rna/rna_multisample"
+  repository:
+    type: "local"
+- name: "workflows/multiomics/dimensionality_reduction"
+  alias: "dimensionality_reduction_rna"
+  repository:
+    type: "local"
+- name: "workflows/multiomics/dimensionality_reduction"
+  alias: "dimensionality_reduction_prot"
+  repository:
+    type: "local"
+- name: "workflows/multiomics/dimensionality_reduction"
+  alias: "dimensionality_reduction_scaling_rna"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/multiomics/process_batches/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/multiomics/process_batches"
+  executable: "target/nextflow/workflows/multiomics/process_batches/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/dataflow/merge"
+  - "target/_private/nextflow/workflows/multiomics/split_modalities"
+  - "target/nextflow/workflows/prot/prot_multisample"
+  - "target/nextflow/workflows/rna/rna_multisample"
+  - "target/nextflow/workflows/multiomics/dimensionality_reduction"
+  - "target/nextflow/workflows/multiomics/dimensionality_reduction"
+  - "target/nextflow/workflows/multiomics/dimensionality_reduction"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/main.nf
new file mode 100644
index 0000000..c5b7616
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/main.nf
@@ -0,0 +1,3945 @@
+// process_batches v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author, maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "process_batches",
+  "namespace" : "workflows/multiomics",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the sample.",
+          "example" : [
+            "foo"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to the sample.",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--rna_layer",
+          "description" : "Input layer for the gene expression modality. If not specified, .X is used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--prot_layer",
+          "description" : "Input layer for the antibody capture modality. If not specified, .X is used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Highly variable features detection",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--highly_variable_features_var_output",
+          "alternatives" : [
+            "--filter_with_hvg_var_output"
+          ],
+          "description" : "In which .var slot to store a boolean array corresponding to the highly variable genes.",
+          "default" : [
+            "filter_with_hvg"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--highly_variable_features_obs_batch_key",
+          "alternatives" : [
+            "--filter_with_hvg_obs_batch_key"
+          ],
+          "description" : "If specified, highly-variable genes are selected within each batch separately and merged. This simple \nprocess avoids the selection of batch-specific genes and acts as a lightweight batch correction method.\n",
+          "default" : [
+            "sample_id"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "QC metrics calculation options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_qc_metrics",
+          "description" : "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes.\n",
+          "example" : [
+            "ercc,highly_variable"
+          ],
+          "default" : [
+            "filter_with_hvg"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        },
+        {
+          "type" : "integer",
+          "name" : "--top_n_vars",
+          "description" : "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated. `--top_n_vars 20,50` finds\ncumulative proportion to the 20th and 50th most expressed vars.\n",
+          "default" : [
+            50,
+            100,
+            200,
+            500
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        }
+      ]
+    },
+    {
+      "name" : "PCA options",
+      "arguments" : [
+        {
+          "type" : "boolean_true",
+          "name" : "--pca_overwrite",
+          "description" : "Allow overwriting slots for PCA output.",
+          "direction" : "input"
+        }
+      ]
+    },
+    {
+      "name" : "CLR options",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--clr_axis",
+          "description" : "Axis to perform the CLR transformation on.",
+          "default" : [
+            0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "RNA Scaling options",
+      "description" : "Options for enabling scaling of the log-normalized data to unit variance and zero mean.\nThe scaled data will be output a different layer and representation with reduced dimensions\nwill be created and stored in addition to the non-scaled data.\n",
+      "arguments" : [
+        {
+          "type" : "boolean_true",
+          "name" : "--rna_enable_scaling",
+          "description" : "Enable scaling for the RNA modality.",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--rna_scaling_output_layer",
+          "description" : "Output layer where the scaled log-normalized data will be stored.",
+          "default" : [
+            "scaled"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--rna_scaling_pca_obsm_output",
+          "description" : "Name of the .obsm key where the PCA representation of the log-normalized\nand scaled data is stored.\n",
+          "default" : [
+            "scaled_pca"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--rna_scaling_pca_loadings_varm_output",
+          "description" : "Name of the .varm key where the PCA loadings of the log-normalized and scaled\ndata is stored.\n",
+          "default" : [
+            "scaled_pca_loadings"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--rna_scaling_pca_variance_uns_output",
+          "description" : "Name of the .uns key where the variance and variance ratio will be stored as a map.\nThe map will contain two keys: variance and variance_ratio respectively.\n",
+          "default" : [
+            "scaled_pca_variance"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--rna_scaling_umap_obsm_output",
+          "description" : "Name of the .obsm key where the UMAP representation of the log-normalized and scaled data is stored.",
+          "default" : [
+            "scaled_umap"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--rna_scaling_max_value",
+          "description" : "Clip (truncate) data to this value after scaling. If not specified, do not clip.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_false",
+          "name" : "--rna_scaling_zero_center",
+          "description" : "If set, omit zero-centering variables, which allows to handle sparse input efficiently.\\"",
+          "direction" : "input"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "This workflow serves as an entrypoint into the 'full_pipeline' in order to\nre-run the multisample processing and the integration setup. An input .h5mu file will \nfirst be split in order to run the multisample processing per modality. Next, the modalities\nare merged again and the integration setup pipeline is executed. Please note that this workflow\nassumes that samples from multiple pipelines are already concatenated. \n",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf2"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_anticmv"
+    }
+  ],
+  "info" : {
+    "test_dependencies" : [
+      {
+        "name" : "workflow_test",
+        "namespace" : "test_workflows/multiomics/process_batches"
+      },
+      {
+        "name" : "workflow_test2",
+        "namespace" : "test_workflows/multiomics/process_batches"
+      }
+    ]
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "dataflow/merge",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/multiomics/split_modalities",
+      "alias" : "split_modalities_workflow",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/prot/prot_multisample",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/rna/rna_multisample",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/multiomics/dimensionality_reduction",
+      "alias" : "dimensionality_reduction_rna",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/multiomics/dimensionality_reduction",
+      "alias" : "dimensionality_reduction_prot",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/multiomics/dimensionality_reduction",
+      "alias" : "dimensionality_reduction_scaling_rna",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/multiomics/process_batches/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/multiomics/process_batches",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { merge } from "${meta.resources_dir}/../../../../nextflow/dataflow/merge/main.nf"
+include { split_modalities as split_modalities_workflow_viashalias } from "${meta.resources_dir}/../../../../_private/nextflow/workflows/multiomics/split_modalities/main.nf"
+split_modalities_workflow = split_modalities_workflow_viashalias.run(key: "split_modalities_workflow")
+include { prot_multisample } from "${meta.resources_dir}/../../../../nextflow/workflows/prot/prot_multisample/main.nf"
+include { rna_multisample } from "${meta.resources_dir}/../../../../nextflow/workflows/rna/rna_multisample/main.nf"
+include { dimensionality_reduction as dimensionality_reduction_rna_viashalias } from "${meta.resources_dir}/../../../../nextflow/workflows/multiomics/dimensionality_reduction/main.nf"
+dimensionality_reduction_rna = dimensionality_reduction_rna_viashalias.run(key: "dimensionality_reduction_rna")
+include { dimensionality_reduction as dimensionality_reduction_prot_viashalias } from "${meta.resources_dir}/../../../../nextflow/workflows/multiomics/dimensionality_reduction/main.nf"
+dimensionality_reduction_prot = dimensionality_reduction_prot_viashalias.run(key: "dimensionality_reduction_prot")
+include { dimensionality_reduction as dimensionality_reduction_scaling_rna_viashalias } from "${meta.resources_dir}/../../../../nextflow/workflows/multiomics/dimensionality_reduction/main.nf"
+dimensionality_reduction_scaling_rna = dimensionality_reduction_scaling_rna_viashalias.run(key: "dimensionality_reduction_scaling_rna")
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+    multisample_ch = input_ch
+      // Make sure there is not conflict between the output from this workflow
+      // And the output from any of the components
+      | map {id, state ->
+        def new_state = state + ["workflow_output": state.output]
+        [id, new_state]
+      }
+      // The input for this workflow can either be a list of unimodal files
+      // or a single multimodal file. To destingish between the two, the files will be split either way.
+      // For multiple unimodal files, the result before or after splitting is identical.
+      // In both cases, this workflow requires split files.
+      
+      // Split must be called on each item of the input list, so split it into multiple events with unique ids
+      // Unique ids are required to run a component
+      | flatMap {id, state ->
+        if (workflow.stubRun) {
+          return [[id, state]]
+        } 
+        def newEvents = state.input.withIndex().collect{input_file, index -> 
+          def newState = state + ["input": input_file, "original_id": id]
+          ["${id}_${index}", newState]
+        }
+        newEvents
+      }
+      | split_modalities_workflow.run(
+        fromState: {id, state ->
+          [
+            "input": state.input,
+            "id": id
+          ]
+        },
+        toState: [
+          "output": "output", 
+          "output_types": "output_types"
+        ]
+      )
+      // gather the output from split_modalities_workflow
+      // by reading the output csv (the csv contains 1 line per output file)
+      | flatMap {id, state ->
+        def outputDir = state.output
+        def csv = state.output_types.splitCsv(strip: true, sep: ",").findAll{!it[0].startsWith("#")}
+        def header = csv.head()
+        def types = csv.tail().collect { row ->
+            [header, row].transpose().collectEntries()
+        }
+        
+        types.collect{ dat ->
+          def new_id = state.original_id + "_${dat.name}" // Make a unique ID by appending the modality name.
+          def new_data = outputDir.resolve(dat.filename)
+          [ new_id, state + ["input": new_data, modality: dat.name]]
+        }
+      }
+      // Remove arguments from split modalities from state
+      | map {id, state -> 
+        def keysToRemove = ["output_types", "original_id"]
+        def newState = state.findAll{it.key !in keysToRemove}
+        [id, newState]
+      }
+    
+    multisample_ch
+      | toSortedList()
+      | map{all_input ->
+        def ids = all_input.collect({it[0]})
+        assert ids.clone().unique().size() == ids.size(): "Found duplicate modalities in the input."
+      }
+
+    //
+    // Multisample processing
+    //
+    def multisample_arguments = [
+      "rna": [
+        "highly_variable_features_var_output": "highly_variable_features_var_output",
+        "highly_variable_features_obs_batch_key": "highly_variable_features_obs_batch_key",
+        "var_qc_metrics": "var_qc_metrics",
+        "top_n_vars": "top_n_vars",
+        "layer": "rna_layer",
+        "enable_scaling": "rna_enable_scaling",
+        "scaling_output_layer": "rna_scaling_output_layer",
+        "scaling_max_value": "rna_scaling_max_value",
+        "scaling_zero_center": "rna_scaling_zero_center",
+      ],
+      "prot": [
+        "layer": "prot_layer",
+        "clr_axis": "clr_axis",
+      ]
+    ].asImmutable()
+
+    multimodal_ch_known = multisample_ch
+      | runEach(
+        components: [rna_multisample, prot_multisample],
+        filter: { id, state, component ->
+          state.modality + "_multisample" == component.config.name
+        },
+        fromState: { id, state, component -> 
+          def newState = multisample_arguments.get(state.modality).collectEntries{key_, value_ -> 
+            [key_, state[value_]]
+          }
+          newState + ["id": id, "input": state.input]
+        },
+        toState: {id, output, state, component ->
+          def newState = state + ["input": output.output]
+          return newState
+        }
+      )
+
+    multimodal_ch_unknown = multisample_ch
+      | filter { id, state -> state.modality !in multisample_arguments.keySet() }
+    
+    multimodal_ch = multimodal_ch_unknown.mix(multimodal_ch_known)
+      // Remove arguments for multisample processing from state.
+      | map {id, state -> 
+        def keysToRemove = multisample_arguments.inject([]){currentKeys, modality, stateMapping -> 
+          def newKeys = currentKeys + stateMapping.values()
+          return newKeys
+        } 
+        keysToRemove -= ["rna_enable_scaling", "rna_scaling_output_layer"]
+        def newState = state.findAll{it.key !in keysToRemove }
+        [id, newState]
+      }
+      | view {"After multisample processing: $it"}
+
+      //
+      // Merging: joining the observations from all modalities together. Everything in 1 file.
+      //
+
+      // Set the original IDs back into place to use them in groupTuple
+      | map {id, state ->
+        def newEvent = [state.id, state]
+        newEvent
+      }
+      // Group the modalities back together per input sample
+      | groupTuple(by: 0, sort: "hash")
+      | view {"After toSortedList: $it"}
+      | map { id, states ->
+          def new_input = states.collect{it.input}
+          def modalities = states.collect{it.modality}.unique()
+          def other_state_keys = states.inject([].toSet()){ current_keys, state ->
+            def new_keys = current_keys + state.keySet()
+            return new_keys
+          }.minus(["output", "input", "modality"])
+          def new_state = other_state_keys.inject([:]){ old_state, argument_name ->
+            argument_values = states.collect{it.get(argument_name)}.unique()
+            assert argument_values.size() == 1, "Arguments should be the same across modalities. Please report this \
+                                                 as a bug. Argument name: $argument_name, \
+                                                 argument value: $argument_values"
+            def argument_value
+            argument_values.each { argument_value = it }
+            def current_state = old_state + [(argument_name): argument_value]
+            return current_state
+          }
+          [id, new_state + ["input": new_input, "modalities": modalities]]
+      }
+      | view {"Input merge channel: $it"}
+      | merge.run(
+        fromState: ["input": "input"],
+        toState: ["input": "output"],
+      )
+      | view {"After merging processing: $it"}
+
+      // Processing of multi-modal multisample MuData files.
+      // Performs calculations on samples that have *not* been integrated,
+      // and can be considered a "no-integration" workflow.
+      output_ch = [dimensionality_reduction_rna, dimensionality_reduction_scaling_rna, dimensionality_reduction_prot].inject(multimodal_ch){ channel_in, component ->
+        channel_out_integrated = channel_in
+          | component.run(
+            runIf: {id, state ->
+              def reg = state.rna_enable_scaling ? ~/^dimensionality_reduction_(scaling_)?/ : ~/^dimensionality_reduction_/
+              def modality_to_check = component.name - reg
+              state.modalities.contains(modality_to_check)
+            },
+            fromState: { id, state -> 
+              def stateMappings = [
+                "dimensionality_reduction_rna": 
+                  [
+                    "id": id,
+                    "input": state.input,
+                    "layer": "log_normalized",
+                    "modality": "rna",
+                    "var_pca_feature_selection": state.highly_variable_features_var_output, // run PCA on highly variable genes only
+                    "pca_overwrite": state.pca_overwrite,
+                    "output": state.workflow_output,
+                  ],
+                "dimensionality_reduction_scaling_rna":
+                  [
+                    "id": id,
+                    "input": state.input,
+                    "layer": state.rna_scaling_output_layer,
+                    "modality": "rna",
+                    "var_pca_feature_selection": state.highly_variable_features_var_output, // run PCA on highly variable genes only
+                    "pca_overwrite": state.pca_overwrite,
+                    // extra scaling args
+                    "obsm_pca": state.rna_scaling_pca_obsm_output,
+                    "pca_loadings_varm_output": state.rna_scaling_pca_loadings_varm_output,
+                    "pca_variance_uns_output": state.rna_scaling_pca_variance_uns_output,
+                    "pca_overwrite": state.pca_overwrite,
+                    "obsm_umap": state.rna_scaling_umap_obsm_output,
+                    "uns_neighbors": "neighbors_scaled",
+                    "obsp_neighbor_connectivities": "connectivities_scaled",
+                    "obsp_neighbor_distances": "distances_scaled",
+                    "output": state.workflow_output,
+                  ],
+                "dimensionality_reduction_prot":
+                  [
+                    "id": id,
+                    "input": state.input,
+                    "layer": "clr",
+                    "modality": "prot",
+                    "pca_overwrite": state.pca_overwrite,
+                    "output": state.workflow_output,
+                  ]
+              ]
+              return stateMappings[component.name]
+            },
+            toState: ["input": "output"]
+          )
+      }
+      // At the end of the reduce statement,
+      // the `toState` closure put the output back into
+      // into the 'input' slot
+      | map {id, state ->
+        [id, ["output": state.input]]
+      }
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/nextflow.config
new file mode 100644
index 0000000..24cdb6b
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/multiomics/process_batches'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'This workflow serves as an entrypoint into the \'full_pipeline\' in order to\nre-run the multisample processing and the integration setup. An input .h5mu file will \nfirst be split in order to run the multisample processing per modality. Next, the modalities\nare merged again and the integration setup pipeline is executed. Please note that this workflow\nassumes that samples from multiple pipelines are already concatenated. \n'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/nextflow_schema.json
new file mode 100644
index 0000000..c831375
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/nextflow_schema.json
@@ -0,0 +1,223 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "process_batches",
+  "description": "This workflow serves as an entrypoint into the 'full_pipeline' in order to\nre-run the multisample processing and the integration setup. An input .h5mu file will \nfirst be split in order to run the multisample processing per modality. Next, the modalities\nare merged again and the integration setup pipeline is executed. Please note that this workflow\nassumes that samples from multiple pipelines are already concatenated. \n",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "ID of the sample.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"foo\"`. "
+        },
+        "input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "exists": true,
+          "description": "Path to the sample.",
+          "help_text": "Type: `file`, multiple: `True`, required, direction: `input`, example: `[\"input.h5mu\"]`. "
+        },
+        "rna_layer": {
+          "type": "string",
+          "description": "Input layer for the gene expression modality",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "prot_layer": {
+          "type": "string",
+          "description": "Input layer for the antibody capture modality",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "highly variable features detection": {
+      "title": "Highly variable features detection",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "highly_variable_features_var_output": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding to the highly variable genes.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"filter_with_hvg\"`. ",
+          "default": "filter_with_hvg"
+        },
+        "highly_variable_features_obs_batch_key": {
+          "type": "string",
+          "description": "If specified, highly-variable genes are selected within each batch separately and merged",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"sample_id\"`. ",
+          "default": "sample_id"
+        }
+      }
+    },
+    "qc metrics calculation options": {
+      "title": "QC metrics calculation options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_qc_metrics": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes.\n",
+          "help_text": "Type: `string`, multiple: `True`, default: `[\"filter_with_hvg\"]`, example: `[\"ercc,highly_variable\"]`. ",
+          "default": [
+            "filter_with_hvg"
+          ]
+        },
+        "top_n_vars": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated",
+          "help_text": "Type: `integer`, multiple: `True`, default: `[50,100,200,500]`. ",
+          "default": [
+            50,
+            100,
+            200,
+            500
+          ]
+        }
+      }
+    },
+    "pca options": {
+      "title": "PCA options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "pca_overwrite": {
+          "type": "boolean",
+          "description": "Allow overwriting slots for PCA output.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "clr options": {
+      "title": "CLR options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "clr_axis": {
+          "type": "integer",
+          "description": "Axis to perform the CLR transformation on.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `0`. ",
+          "default": 0
+        }
+      }
+    },
+    "rna scaling options": {
+      "title": "RNA Scaling options",
+      "type": "object",
+      "description": "Options for enabling scaling of the log-normalized data to unit variance and zero mean.\nThe scaled data will be output a different layer and representation with reduced dimensions\nwill be created and stored in addition to the non-scaled data.\n",
+      "properties": {
+        "rna_enable_scaling": {
+          "type": "boolean",
+          "description": "Enable scaling for the RNA modality.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "rna_scaling_output_layer": {
+          "type": "string",
+          "description": "Output layer where the scaled log-normalized data will be stored.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scaled\"`. ",
+          "default": "scaled"
+        },
+        "rna_scaling_pca_obsm_output": {
+          "type": "string",
+          "description": "Name of the .obsm key where the PCA representation of the log-normalized\nand scaled data is stored.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scaled_pca\"`. ",
+          "default": "scaled_pca"
+        },
+        "rna_scaling_pca_loadings_varm_output": {
+          "type": "string",
+          "description": "Name of the .varm key where the PCA loadings of the log-normalized and scaled\ndata is stored.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scaled_pca_loadings\"`. ",
+          "default": "scaled_pca_loadings"
+        },
+        "rna_scaling_pca_variance_uns_output": {
+          "type": "string",
+          "description": "Name of the .uns key where the variance and variance ratio will be stored as a map.\nThe map will contain two keys: variance and variance_ratio respectively.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scaled_pca_variance\"`. ",
+          "default": "scaled_pca_variance"
+        },
+        "rna_scaling_umap_obsm_output": {
+          "type": "string",
+          "description": "Name of the .obsm key where the UMAP representation of the log-normalized and scaled data is stored.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scaled_umap\"`. ",
+          "default": "scaled_umap"
+        },
+        "rna_scaling_max_value": {
+          "type": "number",
+          "description": "Clip (truncate) data to this value after scaling",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "rna_scaling_zero_center": {
+          "type": "boolean",
+          "description": "If set, omit zero-centering variables, which allows to handle sparse input efficiently.\"",
+          "help_text": "Type: `boolean_false`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/highly variable features detection"
+    },
+    {
+      "$ref": "#/$defs/qc metrics calculation options"
+    },
+    {
+      "$ref": "#/$defs/pca options"
+    },
+    {
+      "$ref": "#/$defs/clr options"
+    },
+    {
+      "$ref": "#/$defs/rna scaling options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_batches/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/.config.vsh.yaml
new file mode 100644
index 0000000..715bf37
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/.config.vsh.yaml
@@ -0,0 +1,768 @@
+name: "process_samples"
+namespace: "workflows/multiomics"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the sample."
+    info: null
+    example:
+    - "foo"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to the sample."
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--rna_layer"
+    description: "Input layer for the gene expression modality. If not specified,\
+      \ .X is used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--prot_layer"
+    description: "Input layer for the antibody capture modality. If not specified,\
+      \ .X is used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--gdo_layer"
+    description: "Input layer for the guide-derived oligonucleotide (GDO) data. If\
+      \ not specified, .X is used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Sample ID options"
+  description: "Options for adding the id to .obs on the MuData object. Having a sample\
+    \ \nid present in a requirement of several components for this pipeline.\n"
+  arguments:
+  - type: "boolean"
+    name: "--add_id_to_obs"
+    description: "Add the value passed with --id to .obs."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--add_id_obs_output"
+    description: ".Obs column to add the sample IDs to. Required and only used when\
+      \ \n--add_id_to_obs is set to 'true'\n"
+    info: null
+    default:
+    - "sample_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--add_id_make_observation_keys_unique"
+    description: "Join the id to the .obs index (.obs_names). \nOnly used when --add_id_to_obs\
+      \ is set to 'true'.\n"
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "RNA filtering options"
+  arguments:
+  - type: "integer"
+    name: "--rna_min_counts"
+    description: "Minimum number of counts captured per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--rna_max_counts"
+    description: "Maximum number of counts captured per cell."
+    info: null
+    example:
+    - 5000000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--rna_min_genes_per_cell"
+    description: "Minimum of non-zero values per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--rna_max_genes_per_cell"
+    description: "Maximum of non-zero values per cell."
+    info: null
+    example:
+    - 1500000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--rna_min_cells_per_gene"
+    description: "Minimum of non-zero values per gene."
+    info: null
+    example:
+    - 3
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--rna_min_fraction_mito"
+    description: "Minimum fraction of UMIs that are mitochondrial."
+    info: null
+    example:
+    - 0.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--rna_max_fraction_mito"
+    description: "Maximum fraction of UMIs that are mitochondrial."
+    info: null
+    example:
+    - 0.2
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--rna_min_fraction_ribo"
+    description: "Minimum fraction of UMIs that are mitochondrial."
+    info: null
+    example:
+    - 0.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--rna_max_fraction_ribo"
+    description: "Maximum fraction of UMIs that are mitochondrial."
+    info: null
+    example:
+    - 0.2
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--skip_scrublet_doublet_detection"
+    description: "Skip the scrublet doublet detection step."
+    info: null
+    direction: "input"
+- name: "CITE-seq filtering options"
+  arguments:
+  - type: "integer"
+    name: "--prot_min_counts"
+    description: "Minimum number of counts per cell."
+    info: null
+    example:
+    - 3
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--prot_max_counts"
+    description: "Minimum number of counts per cell."
+    info: null
+    example:
+    - 5000000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--prot_min_proteins_per_cell"
+    description: "Minimum of non-zero values per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--prot_max_proteins_per_cell"
+    description: "Maximum of non-zero values per cell."
+    info: null
+    example:
+    - 100000000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--prot_min_cells_per_protein"
+    description: "Minimum of non-zero values per protein."
+    info: null
+    example:
+    - 3
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "GDO filtering options"
+  arguments:
+  - type: "integer"
+    name: "--gdo_min_counts"
+    description: "Minimum number of counts per cell."
+    info: null
+    example:
+    - 3
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gdo_max_counts"
+    description: "Minimum number of counts per cell."
+    info: null
+    example:
+    - 5000000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gdo_min_guides_per_cell"
+    description: "Minimum of non-zero values per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gdo_max_guides_per_cell"
+    description: "Maximum of non-zero values per cell."
+    info: null
+    example:
+    - 100000000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gdo_min_cells_per_guide"
+    description: "Minimum of non-zero values per guide."
+    info: null
+    example:
+    - 3
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Highly variable features detection"
+  arguments:
+  - type: "string"
+    name: "--highly_variable_features_var_output"
+    alternatives:
+    - "--filter_with_hvg_var_output"
+    description: "In which .var slot to store a boolean array corresponding to the\
+      \ highly variable genes."
+    info: null
+    default:
+    - "filter_with_hvg"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--highly_variable_features_obs_batch_key"
+    alternatives:
+    - "--filter_with_hvg_obs_batch_key"
+    description: "If specified, highly-variable genes are selected within each batch\
+      \ separately and merged. This simple \nprocess avoids the selection of batch-specific\
+      \ genes and acts as a lightweight batch correction method.\n"
+    info: null
+    default:
+    - "sample_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Mitochondrial & Ribosomal Gene Detection"
+  arguments:
+  - type: "string"
+    name: "--var_gene_names"
+    description: ".var column name to be used to detect mitochondrial/ribosomal genes\
+      \ instead of .var_names (default if not set).\nGene names matching with the\
+      \ regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will\
+      \ be \nidentified as mitochondrial or ribosomal genes, respectively. \n"
+    info: null
+    example:
+    - "gene_symbol"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_mitochondrial_genes"
+    description: "In which .var slot to store a boolean array corresponding the mitochondrial\
+      \ genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_mitochondrial_fraction"
+    description: "When specified, write the fraction of counts originating from mitochondrial\
+      \ genes \n(based on --mitochondrial_gene_regex) to an .obs column with the specified\
+      \ name.\nRequires --var_name_mitochondrial_genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--mitochondrial_gene_regex"
+    description: "Regex string that identifies mitochondrial genes from --var_gene_names.\n\
+      By default will detect human and mouse mitochondrial genes from a gene symbol.\n"
+    info: null
+    default:
+    - "^[mM][tT]-"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_ribosomal_genes"
+    description: "In which .var slot to store a boolean array corresponding the ribosomal\
+      \ genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_ribosomal_fraction"
+    description: "When specified, write the fraction of counts originating from ribosomal\
+      \ genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified\
+      \ name.\nRequires --var_name_ribosomal_genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ribosomal_gene_regex"
+    description: "Regex string that identifies ribosomal genes from --var_gene_names.\n\
+      By default will detect human and mouse ribosomal genes from a gene symbol.\n"
+    info: null
+    default:
+    - "^[Mm]?[Rr][Pp][LlSs]"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "QC metrics calculation options"
+  arguments:
+  - type: "string"
+    name: "--var_qc_metrics"
+    description: "Keys to select a boolean (containing only True or False) column\
+      \ from .var.\nFor each cell, calculate the proportion of total values for genes\
+      \ which are labeled 'True', \ncompared to the total sum of the values for all\
+      \ genes. Defaults to the combined values specified for\n--var_name_mitochondrial_genes\
+      \ and --highly_variable_features_var_output.\n"
+    info: null
+    example:
+    - "ercc,highly_variable"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+  - type: "integer"
+    name: "--top_n_vars"
+    description: "Number of top vars to be used to calculate cumulative proportions.\n\
+      If not specified, proportions are not calculated. `--top_n_vars 20,50` finds\n\
+      cumulative proportion to the 20th and 50th most expressed vars.\n"
+    info: null
+    default:
+    - 50
+    - 100
+    - 200
+    - 500
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+- name: "PCA options"
+  arguments:
+  - type: "boolean_true"
+    name: "--pca_overwrite"
+    description: "Allow overwriting slots for PCA output."
+    info: null
+    direction: "input"
+- name: "CLR options"
+  arguments:
+  - type: "integer"
+    name: "--clr_axis"
+    description: "Axis to perform the CLR transformation on."
+    info: null
+    default:
+    - 0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "RNA Scaling options"
+  description: "Options for enabling scaling of the log-normalized data to unit variance\
+    \ and zero mean.\nThe scaled data will be output a different layer and representation\
+    \ with reduced dimensions\nwill be created and stored in addition to the non-scaled\
+    \ data.\n"
+  arguments:
+  - type: "boolean_true"
+    name: "--rna_enable_scaling"
+    description: "Enable scaling for the RNA modality."
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--rna_scaling_output_layer"
+    description: "Output layer where the scaled log-normalized data will be stored."
+    info: null
+    default:
+    - "scaled"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--rna_scaling_pca_obsm_output"
+    description: "Name of the .obsm key where the PCA representation of the log-normalized\n\
+      and scaled data is stored.\n"
+    info: null
+    default:
+    - "scaled_pca"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--rna_scaling_pca_loadings_varm_output"
+    description: "Name of the .varm key where the PCA loadings of the log-normalized\
+      \ and scaled\ndata is stored.\n"
+    info: null
+    default:
+    - "scaled_pca_loadings"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--rna_scaling_pca_variance_uns_output"
+    description: "Name of the .uns key where the variance and variance ratio will\
+      \ be stored as a map.\nThe map will contain two keys: variance and variance_ratio\
+      \ respectively.\n"
+    info: null
+    default:
+    - "scaled_pca_variance"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--rna_scaling_umap_obsm_output"
+    description: "Name of the .obsm key where the UMAP representation of the log-normalized\
+      \ and scaled data is stored."
+    info: null
+    default:
+    - "scaled_umap"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--rna_scaling_max_value"
+    description: "Clip (truncate) data to this value after scaling. If not specified,\
+      \ do not clip."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_false"
+    name: "--rna_scaling_zero_center"
+    description: "If set, omit zero-centering variables, which allows to handle sparse\
+      \ input efficiently.\""
+    info: null
+    direction: "input"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "A pipeline to analyse multiple multiomics samples."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf2"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf3"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf4"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf5"
+- type: "file"
+  path: "concat_test_data"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+- type: "file"
+  path: "10x_5k_lung_crispr"
+info:
+  test_dependencies:
+  - name: "move_layer"
+    namespace: "transform"
+  - name: "remove_modality"
+    namespace: "filter"
+  - name: "assert_test_workflow_2_output"
+    namespace: "test_workflows/multiomics/process_samples"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "metadata/add_id"
+  repository:
+    type: "local"
+- name: "workflows/multiomics/split_modalities"
+  alias: "split_modalities_workflow"
+  repository:
+    type: "local"
+- name: "dataflow/merge"
+  repository:
+    type: "local"
+- name: "dataflow/concatenate_h5mu"
+  repository:
+    type: "local"
+- name: "workflows/rna/rna_singlesample"
+  repository:
+    type: "local"
+- name: "workflows/prot/prot_singlesample"
+  repository:
+    type: "local"
+- name: "workflows/gdo/gdo_singlesample"
+  repository:
+    type: "local"
+- name: "workflows/multiomics/process_batches"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/multiomics/process_samples/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/multiomics/process_samples"
+  executable: "target/nextflow/workflows/multiomics/process_samples/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/metadata/add_id"
+  - "target/_private/nextflow/workflows/multiomics/split_modalities"
+  - "target/nextflow/dataflow/merge"
+  - "target/nextflow/dataflow/concatenate_h5mu"
+  - "target/nextflow/workflows/rna/rna_singlesample"
+  - "target/nextflow/workflows/prot/prot_singlesample"
+  - "target/nextflow/workflows/gdo/gdo_singlesample"
+  - "target/nextflow/workflows/multiomics/process_batches"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/main.nf
new file mode 100644
index 0000000..6cddb37
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/main.nf
@@ -0,0 +1,4373 @@
+// process_samples v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author, maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "process_samples",
+  "namespace" : "workflows/multiomics",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the sample.",
+          "example" : [
+            "foo"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to the sample.",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--rna_layer",
+          "description" : "Input layer for the gene expression modality. If not specified, .X is used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--prot_layer",
+          "description" : "Input layer for the antibody capture modality. If not specified, .X is used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--gdo_layer",
+          "description" : "Input layer for the guide-derived oligonucleotide (GDO) data. If not specified, .X is used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Sample ID options",
+      "description" : "Options for adding the id to .obs on the MuData object. Having a sample \nid present in a requirement of several components for this pipeline.\n",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--add_id_to_obs",
+          "description" : "Add the value passed with --id to .obs.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--add_id_obs_output",
+          "description" : ".Obs column to add the sample IDs to. Required and only used when \n--add_id_to_obs is set to 'true'\n",
+          "default" : [
+            "sample_id"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--add_id_make_observation_keys_unique",
+          "description" : "Join the id to the .obs index (.obs_names). \nOnly used when --add_id_to_obs is set to 'true'.\n",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "RNA filtering options",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--rna_min_counts",
+          "description" : "Minimum number of counts captured per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--rna_max_counts",
+          "description" : "Maximum number of counts captured per cell.",
+          "example" : [
+            5000000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--rna_min_genes_per_cell",
+          "description" : "Minimum of non-zero values per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--rna_max_genes_per_cell",
+          "description" : "Maximum of non-zero values per cell.",
+          "example" : [
+            1500000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--rna_min_cells_per_gene",
+          "description" : "Minimum of non-zero values per gene.",
+          "example" : [
+            3
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--rna_min_fraction_mito",
+          "description" : "Minimum fraction of UMIs that are mitochondrial.",
+          "example" : [
+            0.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--rna_max_fraction_mito",
+          "description" : "Maximum fraction of UMIs that are mitochondrial.",
+          "example" : [
+            0.2
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--rna_min_fraction_ribo",
+          "description" : "Minimum fraction of UMIs that are mitochondrial.",
+          "example" : [
+            0.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--rna_max_fraction_ribo",
+          "description" : "Maximum fraction of UMIs that are mitochondrial.",
+          "example" : [
+            0.2
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--skip_scrublet_doublet_detection",
+          "description" : "Skip the scrublet doublet detection step.",
+          "direction" : "input"
+        }
+      ]
+    },
+    {
+      "name" : "CITE-seq filtering options",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--prot_min_counts",
+          "description" : "Minimum number of counts per cell.",
+          "example" : [
+            3
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--prot_max_counts",
+          "description" : "Minimum number of counts per cell.",
+          "example" : [
+            5000000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--prot_min_proteins_per_cell",
+          "description" : "Minimum of non-zero values per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--prot_max_proteins_per_cell",
+          "description" : "Maximum of non-zero values per cell.",
+          "example" : [
+            100000000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--prot_min_cells_per_protein",
+          "description" : "Minimum of non-zero values per protein.",
+          "example" : [
+            3
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "GDO filtering options",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--gdo_min_counts",
+          "description" : "Minimum number of counts per cell.",
+          "example" : [
+            3
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gdo_max_counts",
+          "description" : "Minimum number of counts per cell.",
+          "example" : [
+            5000000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gdo_min_guides_per_cell",
+          "description" : "Minimum of non-zero values per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gdo_max_guides_per_cell",
+          "description" : "Maximum of non-zero values per cell.",
+          "example" : [
+            100000000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gdo_min_cells_per_guide",
+          "description" : "Minimum of non-zero values per guide.",
+          "example" : [
+            3
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Highly variable features detection",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--highly_variable_features_var_output",
+          "alternatives" : [
+            "--filter_with_hvg_var_output"
+          ],
+          "description" : "In which .var slot to store a boolean array corresponding to the highly variable genes.",
+          "default" : [
+            "filter_with_hvg"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--highly_variable_features_obs_batch_key",
+          "alternatives" : [
+            "--filter_with_hvg_obs_batch_key"
+          ],
+          "description" : "If specified, highly-variable genes are selected within each batch separately and merged. This simple \nprocess avoids the selection of batch-specific genes and acts as a lightweight batch correction method.\n",
+          "default" : [
+            "sample_id"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Mitochondrial & Ribosomal Gene Detection",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_gene_names",
+          "description" : ".var column name to be used to detect mitochondrial/ribosomal genes instead of .var_names (default if not set).\nGene names matching with the regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will be \nidentified as mitochondrial or ribosomal genes, respectively. \n",
+          "example" : [
+            "gene_symbol"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_mitochondrial_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_mitochondrial_fraction",
+          "description" : "When specified, write the fraction of counts originating from mitochondrial genes \n(based on --mitochondrial_gene_regex) to an .obs column with the specified name.\nRequires --var_name_mitochondrial_genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--mitochondrial_gene_regex",
+          "description" : "Regex string that identifies mitochondrial genes from --var_gene_names.\nBy default will detect human and mouse mitochondrial genes from a gene symbol.\n",
+          "default" : [
+            "^[mM][tT]-"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_ribosomal_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_ribosomal_fraction",
+          "description" : "When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified name.\nRequires --var_name_ribosomal_genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--ribosomal_gene_regex",
+          "description" : "Regex string that identifies ribosomal genes from --var_gene_names.\nBy default will detect human and mouse ribosomal genes from a gene symbol.\n",
+          "default" : [
+            "^[Mm]?[Rr][Pp][LlSs]"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "QC metrics calculation options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_qc_metrics",
+          "description" : "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes. Defaults to the combined values specified for\n--var_name_mitochondrial_genes and --highly_variable_features_var_output.\n",
+          "example" : [
+            "ercc,highly_variable"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        },
+        {
+          "type" : "integer",
+          "name" : "--top_n_vars",
+          "description" : "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated. `--top_n_vars 20,50` finds\ncumulative proportion to the 20th and 50th most expressed vars.\n",
+          "default" : [
+            50,
+            100,
+            200,
+            500
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        }
+      ]
+    },
+    {
+      "name" : "PCA options",
+      "arguments" : [
+        {
+          "type" : "boolean_true",
+          "name" : "--pca_overwrite",
+          "description" : "Allow overwriting slots for PCA output.",
+          "direction" : "input"
+        }
+      ]
+    },
+    {
+      "name" : "CLR options",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--clr_axis",
+          "description" : "Axis to perform the CLR transformation on.",
+          "default" : [
+            0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "RNA Scaling options",
+      "description" : "Options for enabling scaling of the log-normalized data to unit variance and zero mean.\nThe scaled data will be output a different layer and representation with reduced dimensions\nwill be created and stored in addition to the non-scaled data.\n",
+      "arguments" : [
+        {
+          "type" : "boolean_true",
+          "name" : "--rna_enable_scaling",
+          "description" : "Enable scaling for the RNA modality.",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--rna_scaling_output_layer",
+          "description" : "Output layer where the scaled log-normalized data will be stored.",
+          "default" : [
+            "scaled"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--rna_scaling_pca_obsm_output",
+          "description" : "Name of the .obsm key where the PCA representation of the log-normalized\nand scaled data is stored.\n",
+          "default" : [
+            "scaled_pca"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--rna_scaling_pca_loadings_varm_output",
+          "description" : "Name of the .varm key where the PCA loadings of the log-normalized and scaled\ndata is stored.\n",
+          "default" : [
+            "scaled_pca_loadings"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--rna_scaling_pca_variance_uns_output",
+          "description" : "Name of the .uns key where the variance and variance ratio will be stored as a map.\nThe map will contain two keys: variance and variance_ratio respectively.\n",
+          "default" : [
+            "scaled_pca_variance"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--rna_scaling_umap_obsm_output",
+          "description" : "Name of the .obsm key where the UMAP representation of the log-normalized and scaled data is stored.",
+          "default" : [
+            "scaled_umap"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--rna_scaling_max_value",
+          "description" : "Clip (truncate) data to this value after scaling. If not specified, do not clip.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_false",
+          "name" : "--rna_scaling_zero_center",
+          "description" : "If set, omit zero-centering variables, which allows to handle sparse input efficiently.\\"",
+          "direction" : "input"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "A pipeline to analyse multiple multiomics samples.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf2"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf3"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf4"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf5"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_lung_crispr"
+    }
+  ],
+  "info" : {
+    "test_dependencies" : [
+      {
+        "name" : "move_layer",
+        "namespace" : "transform"
+      },
+      {
+        "name" : "remove_modality",
+        "namespace" : "filter"
+      },
+      {
+        "name" : "assert_test_workflow_2_output",
+        "namespace" : "test_workflows/multiomics/process_samples"
+      }
+    ]
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "metadata/add_id",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/multiomics/split_modalities",
+      "alias" : "split_modalities_workflow",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "dataflow/merge",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "dataflow/concatenate_h5mu",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/rna/rna_singlesample",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/prot/prot_singlesample",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/gdo/gdo_singlesample",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/multiomics/process_batches",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/multiomics/process_samples/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/multiomics/process_samples",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { add_id } from "${meta.resources_dir}/../../../../nextflow/metadata/add_id/main.nf"
+include { split_modalities as split_modalities_workflow_viashalias } from "${meta.resources_dir}/../../../../_private/nextflow/workflows/multiomics/split_modalities/main.nf"
+split_modalities_workflow = split_modalities_workflow_viashalias.run(key: "split_modalities_workflow")
+include { merge } from "${meta.resources_dir}/../../../../nextflow/dataflow/merge/main.nf"
+include { concatenate_h5mu } from "${meta.resources_dir}/../../../../nextflow/dataflow/concatenate_h5mu/main.nf"
+include { rna_singlesample } from "${meta.resources_dir}/../../../../nextflow/workflows/rna/rna_singlesample/main.nf"
+include { prot_singlesample } from "${meta.resources_dir}/../../../../nextflow/workflows/prot/prot_singlesample/main.nf"
+include { gdo_singlesample } from "${meta.resources_dir}/../../../../nextflow/workflows/gdo/gdo_singlesample/main.nf"
+include { process_batches } from "${meta.resources_dir}/../../../../nextflow/workflows/multiomics/process_batches/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+    input_ch
+
+  main:
+    modalities_ch = input_ch
+      // Make sure there is not conflict between the output from this workflow
+      // And the output from any of the components
+      | map {id, state ->
+        def new_state = state + ["workflow_output": state.output]
+        [id, new_state]
+      }
+      // If requested to be detected, make sure the mitochondrial and ribosomal genes
+      // are added to the input of the qc metrics calculation
+      | map {id, state ->
+        def var_qc_default = [state.highly_variable_features_var_output]
+        if (state.var_name_mitochondrial_genes) {
+          var_qc_default.add(state.var_name_mitochondrial_genes)
+        }
+        if (state.var_name_ribosomal_genes) {
+          var_qc_default.add(state.var_name_ribosomal_genes)
+        }
+        def newState = state + ["var_qc_metrics": var_qc_default.join(",")]
+        [id, newState]
+      }
+      // If requested, add the id of the events (samples) a column in .obs. 
+      // Also allows to make .obs_names (the .obs index) unique, by prefixing the values with an unique id per .h5mu file.
+      // The latter is usefull to avoid duplicate observations during concatenation.
+      | add_id.run(
+        filter: {id, state -> state.add_id_to_obs },
+        fromState: {id, state -> 
+          def newState = [
+            "input": state.input,
+            "input_id": id,
+            "make_observation_keys_unique": state.add_id_make_observation_keys_unique,
+            "obs_output": state.add_id_obs_output,
+            "add_id_to_obs": state.add_id_to_obs
+          ]
+          newState
+        },
+        toState: {id, output, state ->
+          def keysToRemove = ["add_id_to_obs", "add_id_obs_output", "add_id_make_observation_keys_unique"]
+          def newState = state.findAll{it.key !in keysToRemove}
+          newState + ["input": output.output]
+        }
+      )
+      | split_modalities_workflow.run(
+        fromState: {id, state ->
+          def newState = ["input": state.input, "id": id]
+        },
+        toState: ["output": "output", "output_types": "output_types"]
+      )
+      | flatMap {id, state ->
+        def outputDir = state.output
+        def csv = state.output_types.splitCsv(strip: true, sep: ",").findAll{!it[0].startsWith("#")}
+        def header = csv.head()
+        def types = csv.tail().collect { row ->
+            [header, row].transpose().collectEntries()
+        }
+        
+        types.collect{ dat ->
+          // def new_id = id + "_" + dat.name
+          def new_id = id // it's okay because the channel will get split up anyways
+          def new_data = outputDir.resolve(dat.filename)
+          [ new_id, state + ["input": new_data, modality: dat.name]]
+        }
+      }
+      // Remove arguments from split modalities from state
+      | map {id, state -> 
+        def keysToRemove = ["output_types"]
+        def newState = state.findAll{it.key !in keysToRemove}
+        [id, newState]
+      }
+      | view {"After splitting modalities: $it"}
+
+
+    //
+    // Singlesample processing
+    //
+    def singlesample_arguments = [
+      "rna": [
+        "min_counts": "rna_min_counts",
+        "max_counts": "rna_max_counts",
+        "min_genes_per_cell": "rna_min_genes_per_cell",
+        "max_genes_per_cell": "rna_max_genes_per_cell",
+        "min_cells_per_gene": "rna_min_cells_per_gene",
+        "min_fraction_mito": "rna_min_fraction_mito",
+        "max_fraction_mito": "rna_max_fraction_mito",
+        "var_name_mitochondrial_genes": "var_name_mitochondrial_genes",
+        "obs_name_mitochondrial_fraction": "obs_name_mitochondrial_fraction",
+        "var_gene_names": "var_gene_names",
+        "mitochondrial_gene_regex": "mitochondrial_gene_regex",
+        "min_fraction_ribo": "rna_min_fraction_ribo",
+        "max_fraction_ribo": "rna_max_fraction_ribo",
+        "var_name_ribosomal_genes": "var_name_ribosomal_genes",
+        "obs_name_ribosomal_fraction": "obs_name_ribosomal_fraction",
+        "ribosomal_gene_regex": "ribosomal_gene_regex",
+        "layer": "rna_layer",
+        "skip_scrublet_doublet_detection": "skip_scrublet_doublet_detection"
+      ],
+      "prot": [
+        "min_counts": "prot_min_counts",
+        "max_counts": "prot_max_counts",
+        "min_proteins_per_cell": "prot_min_proteins_per_cell",
+        "max_proteins_per_cell": "prot_max_proteins_per_cell",
+        "min_cells_per_protein": "prot_min_cells_per_protein",
+        "layer": "prot_layer",
+      ],
+      "gdo": [
+        "min_counts": "gdo_min_counts",
+        "max_counts": "gdo_max_counts",
+        "min_guides_per_cell": "gdo_min_guides_per_cell",
+        "max_guides_per_cell": "gdo_max_guides_per_cell",
+        "min_cells_per_guide": "gdo_min_cells_per_guide",
+        "layer": "gdo_layer",
+      ], 
+    ].asImmutable()
+
+    multisample_ch_known = modalities_ch 
+      // run the singlesample processing
+      | runEach(
+        components: [rna_singlesample, prot_singlesample, gdo_singlesample],
+        filter: { id, state, component ->
+          state.modality + "_singlesample" == component.config.name
+        },
+        fromState: { id, state, component ->
+          def newState = singlesample_arguments.get(state.modality).collectEntries{key_, value_ -> 
+            [key_, state[value_]]
+          }
+          return newState + ["id": id, "input": state.input]
+        },
+        toState: ["input": "output"],
+      )
+
+    multisample_ch_unknown = modalities_ch
+      | filter{id, state -> state.modality !in singlesample_arguments.keySet()}
+
+    output_ch = multisample_ch_unknown.mix(multisample_ch_known)
+      // Remove arguments for singlesample processing from state.
+      | map {id, state -> 
+        def keysToRemove = singlesample_arguments.inject([]){currentKeys, modality, stateMapping -> 
+            currentKeys += stateMapping.values()
+        }
+        def allwayskeep = ["gdo_layer", "rna_layer", "prot_layer", "workflow_output"]
+        def newState = state.findAll{(it.key !in keysToRemove + ["id"]) || (it.key in allwayskeep)}
+        [id, newState]
+      }
+      | view {"After singlesample processing: $it"}
+
+      //
+      // Concatenation: join observations across samples together per modality. 
+      //
+      // Concatenate multiple single-sample unimodal MuData objects back into several multi-sample files.
+      // One multi-sample MuData file is created per modality.
+      //
+      | map { id, state -> // Put modality name in first element so that we can group on it
+        [state.modality, id, state]
+      }
+      | groupTuple(by: 0, sort: "hash")
+      | view {"After groupTuple: $it"}
+      | map { modality, old_ids, states ->
+        def new_id = "combined_$modality"
+        // keys in the new state that should not have a unique value across samples
+        def new_state_non_unique_values = [
+          "input": states.collect{it.input},
+          "input_id": old_ids,
+          "_meta": ["join_id": old_ids[0]]
+        ]
+        // Gather the keys from the different states,
+        // one state might contain more keys compared to another (so create a set)
+        def all_state_keys = states.inject([].toSet()){ current_keys, state ->
+            def new_keys = current_keys + state.keySet()
+            return new_keys
+        }.minus(["output", "input_id", "input", "_meta"])
+        // Create the new state from the keys, values should be the same across samples
+        def new_state = all_state_keys.inject([:]){ old_state, argument_name ->
+            argument_values = states.collect{it.get(argument_name)}.unique()
+            assert argument_values.size() == 1, "Arguments should be the same across samples. Argument name: $argument_name, \
+                                                 argument value: $argument_values"
+            // take the unique value from the set (there is only one)
+            def argument_value
+            argument_values.each { argument_value = it }
+            def current_state = old_state + [(argument_name): argument_value]
+            return current_state
+        }
+        def final_state = new_state_non_unique_values + new_state
+        [new_id, final_state]
+      }
+      | concatenate_h5mu.run(
+        fromState: [
+          "input": "input",
+          "input_id": "input_id"
+        ],
+        toState: {id, output, state -> 
+          def keysToRemove = ["input_id"]
+          def newState = state.findAll{it.key !in keysToRemove}
+          newState + ["input": output.output]
+        }, 
+      )
+
+      | view {"After concatenation: $it"}
+      | toSortedList()
+      | map {modalities_states -> 
+        def states = modalities_states.collect{it[1]}
+        def new_input = states.collect{it.input}
+        def join_id = states[0]._meta.join_id
+        def other_state_keys = states.inject([].toSet()){ current_keys, state ->
+            def new_keys = current_keys + state.keySet()
+            return new_keys
+          }.minus(["output", "input", "modality", "_meta"])
+        def new_state = other_state_keys.inject([:]){ old_state, argument_name ->
+            argument_values = states.collect{it.get(argument_name)}.unique()
+            assert argument_values.size() == 1, "Arguments should be the same across modalities. Please report this \
+                                                 as a bug. Argument name: $argument_name, \
+                                                 argument value: $argument_values"
+            def argument_value
+            argument_values.each { argument_value = it }
+            def current_state = old_state + [(argument_name): argument_value]
+            return current_state
+          }
+         ["merged", new_state + ["input": new_input, "_meta": ["join_id": join_id]]]
+      }
+      | process_batches.run(
+        fromState: {id, state ->
+          [
+            "id": id,
+            "input": state.input,
+            "output": state.workflow_output,
+            "highly_variable_features_var_output": state.highly_variable_features_var_output,
+            "highly_variable_features_obs_batch_key": state.highly_variable_features_obs_batch_key,
+            "var_qc_metrics": state.var_qc_metrics,
+            "top_n_vars": state.top_n_vars, 
+            "pca_overwrite": state.pca_overwrite,
+            "rna_layer": state.rna_layer,
+            "prot_layer": state.prot_layer,
+            "clr_axis": state.clr_axis,
+            "rna_enable_scaling": state.rna_enable_scaling,
+            "rna_scaling_output_layer": state.rna_scaling_output_layer,
+            "rna_scaling_pca_obsm_output": state.rna_scaling_pca_obsm_output,
+            "rna_scaling_pca_loadings_varm_output": state.rna_scaling_pca_loadings_varm_output,
+            "rna_scaling_pca_variance_uns_output": state.rna_scaling_pca_variance_uns_output,
+            "rna_scaling_umap_obsm_output": state.rna_scaling_umap_obsm_output,
+            "rna_scaling_max_value": state.rna_scaling_max_value,
+            "rna_scaling_zero_center": state.rna_scaling_zero_center,
+          ]
+        },
+        toState: {id, output, state -> 
+          [
+            "output": output.output,
+            "_meta": state._meta,
+          ]
+        }
+      )
+      | view {"After process_batches: $it"}
+
+  emit:
+    output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/nextflow.config
new file mode 100644
index 0000000..d56b318
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/multiomics/process_samples'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'A pipeline to analyse multiple multiomics samples.'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/nextflow_schema.json
new file mode 100644
index 0000000..dab191e
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/nextflow_schema.json
@@ -0,0 +1,428 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "process_samples",
+  "description": "A pipeline to analyse multiple multiomics samples.",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "ID of the sample.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"foo\"`. "
+        },
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the sample.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "rna_layer": {
+          "type": "string",
+          "description": "Input layer for the gene expression modality",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "prot_layer": {
+          "type": "string",
+          "description": "Input layer for the antibody capture modality",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "gdo_layer": {
+          "type": "string",
+          "description": "Input layer for the guide-derived oligonucleotide (GDO) data",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "sample id options": {
+      "title": "Sample ID options",
+      "type": "object",
+      "description": "Options for adding the id to .obs on the MuData object. Having a sample \nid present in a requirement of several components for this pipeline.\n",
+      "properties": {
+        "add_id_to_obs": {
+          "type": "boolean",
+          "description": "Add the value passed with --id to .obs.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "add_id_obs_output": {
+          "type": "string",
+          "description": ".Obs column to add the sample IDs to",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"sample_id\"`. ",
+          "default": "sample_id"
+        },
+        "add_id_make_observation_keys_unique": {
+          "type": "boolean",
+          "description": "Join the id to the .obs index (.obs_names)",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "rna filtering options": {
+      "title": "RNA filtering options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "rna_min_counts": {
+          "type": "integer",
+          "description": "Minimum number of counts captured per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "rna_max_counts": {
+          "type": "integer",
+          "description": "Maximum number of counts captured per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `5000000`. "
+        },
+        "rna_min_genes_per_cell": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "rna_max_genes_per_cell": {
+          "type": "integer",
+          "description": "Maximum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `1500000`. "
+        },
+        "rna_min_cells_per_gene": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per gene.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3`. "
+        },
+        "rna_min_fraction_mito": {
+          "type": "number",
+          "description": "Minimum fraction of UMIs that are mitochondrial.",
+          "help_text": "Type: `double`, multiple: `False`, example: `0.0`. "
+        },
+        "rna_max_fraction_mito": {
+          "type": "number",
+          "description": "Maximum fraction of UMIs that are mitochondrial.",
+          "help_text": "Type: `double`, multiple: `False`, example: `0.2`. "
+        },
+        "rna_min_fraction_ribo": {
+          "type": "number",
+          "description": "Minimum fraction of UMIs that are mitochondrial.",
+          "help_text": "Type: `double`, multiple: `False`, example: `0.0`. "
+        },
+        "rna_max_fraction_ribo": {
+          "type": "number",
+          "description": "Maximum fraction of UMIs that are mitochondrial.",
+          "help_text": "Type: `double`, multiple: `False`, example: `0.2`. "
+        },
+        "skip_scrublet_doublet_detection": {
+          "type": "boolean",
+          "description": "Skip the scrublet doublet detection step.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "cite-seq filtering options": {
+      "title": "CITE-seq filtering options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "prot_min_counts": {
+          "type": "integer",
+          "description": "Minimum number of counts per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3`. "
+        },
+        "prot_max_counts": {
+          "type": "integer",
+          "description": "Minimum number of counts per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `5000000`. "
+        },
+        "prot_min_proteins_per_cell": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "prot_max_proteins_per_cell": {
+          "type": "integer",
+          "description": "Maximum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `100000000`. "
+        },
+        "prot_min_cells_per_protein": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per protein.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3`. "
+        }
+      }
+    },
+    "gdo filtering options": {
+      "title": "GDO filtering options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "gdo_min_counts": {
+          "type": "integer",
+          "description": "Minimum number of counts per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3`. "
+        },
+        "gdo_max_counts": {
+          "type": "integer",
+          "description": "Minimum number of counts per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `5000000`. "
+        },
+        "gdo_min_guides_per_cell": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "gdo_max_guides_per_cell": {
+          "type": "integer",
+          "description": "Maximum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `100000000`. "
+        },
+        "gdo_min_cells_per_guide": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per guide.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3`. "
+        }
+      }
+    },
+    "highly variable features detection": {
+      "title": "Highly variable features detection",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "highly_variable_features_var_output": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding to the highly variable genes.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"filter_with_hvg\"`. ",
+          "default": "filter_with_hvg"
+        },
+        "highly_variable_features_obs_batch_key": {
+          "type": "string",
+          "description": "If specified, highly-variable genes are selected within each batch separately and merged",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"sample_id\"`. ",
+          "default": "sample_id"
+        }
+      }
+    },
+    "mitochondrial & ribosomal gene detection": {
+      "title": "Mitochondrial & Ribosomal Gene Detection",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_gene_names": {
+          "type": "string",
+          "description": ".var column name to be used to detect mitochondrial/ribosomal genes instead of .var_names (default if not set).\nGene names matching with the regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will be \nidentified as mitochondrial or ribosomal genes, respectively",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gene_symbol\"`. "
+        },
+        "var_name_mitochondrial_genes": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_name_mitochondrial_fraction": {
+          "type": "string",
+          "description": "When specified, write the fraction of counts originating from mitochondrial genes \n(based on --mitochondrial_gene_regex) to an .obs column with the specified name.\nRequires --var_name_mitochondrial_genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "mitochondrial_gene_regex": {
+          "type": "string",
+          "description": "Regex string that identifies mitochondrial genes from --var_gene_names.\nBy default will detect human and mouse mitochondrial genes from a gene symbol.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"^[mM][tT]-\"`. ",
+          "default": "^[mM][tT]-"
+        },
+        "var_name_ribosomal_genes": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_name_ribosomal_fraction": {
+          "type": "string",
+          "description": "When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified name.\nRequires --var_name_ribosomal_genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "ribosomal_gene_regex": {
+          "type": "string",
+          "description": "Regex string that identifies ribosomal genes from --var_gene_names.\nBy default will detect human and mouse ribosomal genes from a gene symbol.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"^[Mm]?[Rr][Pp][LlSs]\"`. ",
+          "default": "^[Mm]?[Rr][Pp][LlSs]"
+        }
+      }
+    },
+    "qc metrics calculation options": {
+      "title": "QC metrics calculation options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_qc_metrics": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"ercc,highly_variable\"]`. "
+        },
+        "top_n_vars": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated",
+          "help_text": "Type: `integer`, multiple: `True`, default: `[50,100,200,500]`. ",
+          "default": [
+            50,
+            100,
+            200,
+            500
+          ]
+        }
+      }
+    },
+    "pca options": {
+      "title": "PCA options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "pca_overwrite": {
+          "type": "boolean",
+          "description": "Allow overwriting slots for PCA output.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "clr options": {
+      "title": "CLR options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "clr_axis": {
+          "type": "integer",
+          "description": "Axis to perform the CLR transformation on.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `0`. ",
+          "default": 0
+        }
+      }
+    },
+    "rna scaling options": {
+      "title": "RNA Scaling options",
+      "type": "object",
+      "description": "Options for enabling scaling of the log-normalized data to unit variance and zero mean.\nThe scaled data will be output a different layer and representation with reduced dimensions\nwill be created and stored in addition to the non-scaled data.\n",
+      "properties": {
+        "rna_enable_scaling": {
+          "type": "boolean",
+          "description": "Enable scaling for the RNA modality.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "rna_scaling_output_layer": {
+          "type": "string",
+          "description": "Output layer where the scaled log-normalized data will be stored.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scaled\"`. ",
+          "default": "scaled"
+        },
+        "rna_scaling_pca_obsm_output": {
+          "type": "string",
+          "description": "Name of the .obsm key where the PCA representation of the log-normalized\nand scaled data is stored.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scaled_pca\"`. ",
+          "default": "scaled_pca"
+        },
+        "rna_scaling_pca_loadings_varm_output": {
+          "type": "string",
+          "description": "Name of the .varm key where the PCA loadings of the log-normalized and scaled\ndata is stored.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scaled_pca_loadings\"`. ",
+          "default": "scaled_pca_loadings"
+        },
+        "rna_scaling_pca_variance_uns_output": {
+          "type": "string",
+          "description": "Name of the .uns key where the variance and variance ratio will be stored as a map.\nThe map will contain two keys: variance and variance_ratio respectively.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scaled_pca_variance\"`. ",
+          "default": "scaled_pca_variance"
+        },
+        "rna_scaling_umap_obsm_output": {
+          "type": "string",
+          "description": "Name of the .obsm key where the UMAP representation of the log-normalized and scaled data is stored.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scaled_umap\"`. ",
+          "default": "scaled_umap"
+        },
+        "rna_scaling_max_value": {
+          "type": "number",
+          "description": "Clip (truncate) data to this value after scaling",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "rna_scaling_zero_center": {
+          "type": "boolean",
+          "description": "If set, omit zero-centering variables, which allows to handle sparse input efficiently.\"",
+          "help_text": "Type: `boolean_false`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/sample id options"
+    },
+    {
+      "$ref": "#/$defs/rna filtering options"
+    },
+    {
+      "$ref": "#/$defs/cite-seq filtering options"
+    },
+    {
+      "$ref": "#/$defs/gdo filtering options"
+    },
+    {
+      "$ref": "#/$defs/highly variable features detection"
+    },
+    {
+      "$ref": "#/$defs/mitochondrial & ribosomal gene detection"
+    },
+    {
+      "$ref": "#/$defs/qc metrics calculation options"
+    },
+    {
+      "$ref": "#/$defs/pca options"
+    },
+    {
+      "$ref": "#/$defs/clr options"
+    },
+    {
+      "$ref": "#/$defs/rna scaling options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/.config.vsh.yaml
new file mode 100644
index 0000000..4453d54
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/.config.vsh.yaml
@@ -0,0 +1,335 @@
+name: "prot_multisample"
+namespace: "workflows/prot"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the concatenated file"
+    info: null
+    example:
+    - "concatenated"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    description: "Path to the samples."
+    info: null
+    example:
+    - "dataset.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Input layer to use. If not specified, .X is used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "QC metrics calculation options"
+  arguments:
+  - type: "string"
+    name: "--var_qc_metrics"
+    description: "Keys to select a boolean (containing only True or False) column\
+      \ from .var.\nFor each cell, calculate the proportion of total values for genes\
+      \ which are labeled 'True', \ncompared to the total sum of the values for all\
+      \ genes. Defaults to the value from\n--var_name_mitochondrial_genes.\n"
+    info: null
+    example:
+    - "ercc,highly_variable"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+  - type: "integer"
+    name: "--top_n_vars"
+    description: "Number of top vars to be used to calculate cumulative proportions.\n\
+      If not specified, proportions are not calculated. `--top_n_vars 20,50` finds\n\
+      cumulative proportion to the 20th and 50th most expressed vars.\n"
+    info: null
+    default:
+    - 50
+    - 100
+    - 200
+    - 500
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+  - type: "string"
+    name: "--output_obs_num_nonzero_vars"
+    description: "Name of column in .obs describing, for each observation, the number\
+      \ of stored values\n(including explicit zeroes). In other words, the name of\
+      \ the column that counts\nfor each row the number of columns that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_vars"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_total_counts_vars"
+    description: "Name of the column for .obs describing, for each observation (row),\n\
+      the sum of the stored values in the columns.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_num_nonzero_obs"
+    description: "Name of column describing, for each feature, the number of stored\
+      \ values\n(including explicit zeroes). In other words, the name of the column\
+      \ that counts\nfor each column the number of rows that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_obs"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_total_counts_obs"
+    description: "Name of the column in .var describing, for each feature (column),\n\
+      the sum of the stored values in the rows.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_obs_mean"
+    description: "Name of the column in .obs providing the mean of the values in each\
+      \ row.\n"
+    info: null
+    default:
+    - "obs_mean"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_pct_dropout"
+    description: "Name of the column in .obs providing for each feature the percentage\
+      \ of\nobservations the feature does not appear on (i.e. is missing). Same as\
+      \ `--output_var_num_nonzero_obs`\nbut percentage based.\n"
+    info: null
+    default:
+    - "pct_dropout"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "CLR arguments"
+  arguments:
+  - type: "integer"
+    name: "--clr_axis"
+    description: "Axis across which CLR is performed."
+    info: null
+    default:
+    - 0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Processing unimodal multi-sample ADT data."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info:
+  image: "/images/concepts/fig_workflow_multiomics_adt_multisample.svg"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "transform/clr"
+  repository:
+    type: "local"
+- name: "workflows/qc/qc"
+  alias: "prot_qc"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/prot/prot_multisample/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/prot/prot_multisample"
+  executable: "target/nextflow/workflows/prot/prot_multisample/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/transform/clr"
+  - "target/nextflow/workflows/qc/qc"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/main.nf
new file mode 100644
index 0000000..a33395c
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/main.nf
@@ -0,0 +1,3629 @@
+// prot_multisample v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "prot_multisample",
+  "namespace" : "workflows/prot",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the concatenated file",
+          "example" : [
+            "concatenated"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Path to the samples.",
+          "example" : [
+            "dataset.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Input layer to use. If not specified, .X is used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "QC metrics calculation options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_qc_metrics",
+          "description" : "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes. Defaults to the value from\n--var_name_mitochondrial_genes.\n",
+          "example" : [
+            "ercc,highly_variable"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        },
+        {
+          "type" : "integer",
+          "name" : "--top_n_vars",
+          "description" : "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated. `--top_n_vars 20,50` finds\ncumulative proportion to the 20th and 50th most expressed vars.\n",
+          "default" : [
+            50,
+            100,
+            200,
+            500
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_num_nonzero_vars",
+          "description" : "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each row the number of columns that contain data.\n",
+          "default" : [
+            "num_nonzero_vars"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_total_counts_vars",
+          "description" : "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_num_nonzero_obs",
+          "description" : "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each column the number of rows that contain data.\n",
+          "default" : [
+            "num_nonzero_obs"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_total_counts_obs",
+          "description" : "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_obs_mean",
+          "description" : "Name of the column in .obs providing the mean of the values in each row.\n",
+          "default" : [
+            "obs_mean"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_pct_dropout",
+          "description" : "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e. is missing). Same as `--output_var_num_nonzero_obs`\nbut percentage based.\n",
+          "default" : [
+            "pct_dropout"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "CLR arguments",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--clr_axis",
+          "description" : "Axis across which CLR is performed.",
+          "default" : [
+            0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Processing unimodal multi-sample ADT data.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "info" : {
+    "image" : "/images/concepts/fig_workflow_multiomics_adt_multisample.svg"
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "transform/clr",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/qc/qc",
+      "alias" : "prot_qc",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/prot/prot_multisample/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/prot/prot_multisample",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { clr } from "${meta.resources_dir}/../../../../nextflow/transform/clr/main.nf"
+include { qc as prot_qc_viashalias } from "${meta.resources_dir}/../../../../nextflow/workflows/qc/qc/main.nf"
+prot_qc = prot_qc_viashalias.run(key: "prot_qc")
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+
+  output_ch = input_ch
+    | map {id, state ->
+      def new_state = state + ["workflow_output": state.output]
+      [id, new_state]
+    }
+    | clr.run(
+      fromState: [
+        "input": "input",
+        "input_layer": "layer",
+        "clr_axis": "clr_axis",
+        ],
+      toState: ["input": "output"],
+      args: [ 
+        output_layer: "clr", 
+        modality: "prot"
+      ]
+    )
+    | prot_qc.run(
+      // TODO: remove when viash 0.8.3 is released
+      key: "prot_qc",
+      fromState: { id, state ->
+        def newState = [
+          "id": id,
+          "output": state.workflow_output,
+          "input": state.input,
+          "top_n_vars": state.top_n_vars,
+          "var_qc_metrics": null,
+          "input_layer": state.layer, // Use the non-transformed layer
+          "modality": "prot",
+          "var_name_mitochondrial_genes": null,
+          "output_obs_num_nonzero_vars": state.output_obs_num_nonzero_vars,
+          "output_obs_total_counts_vars": state.output_obs_total_counts_vars,
+          "num_nonzeoutput_var_num_nonzero_obsro_obs": state.output_var_num_nonzero_obs,
+          "output_var_total_counts_obs": state.output_var_total_counts_obs,
+          "output_var_obs_mean": state.output_var_obs_mean,
+          "output_var_pct_dropout": state.pct_dropout
+        ]
+        newState
+      }
+    )
+    | setState(["output"])
+
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/nextflow.config
new file mode 100644
index 0000000..4d8af19
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/prot/prot_multisample'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Processing unimodal multi-sample ADT data.'
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/nextflow_schema.json
new file mode 100644
index 0000000..6402741
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/nextflow_schema.json
@@ -0,0 +1,153 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "prot_multisample",
+  "description": "Processing unimodal multi-sample ADT data.",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "ID of the concatenated file",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"concatenated\"`. "
+        },
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the samples.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"dataset.h5mu\"`. "
+        },
+        "layer": {
+          "type": "string",
+          "description": "Input layer to use",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "qc metrics calculation options": {
+      "title": "QC metrics calculation options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_qc_metrics": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"ercc,highly_variable\"]`. "
+        },
+        "top_n_vars": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated",
+          "help_text": "Type: `integer`, multiple: `True`, default: `[50,100,200,500]`. ",
+          "default": [
+            50,
+            100,
+            200,
+            500
+          ]
+        },
+        "output_obs_num_nonzero_vars": {
+          "type": "string",
+          "description": "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes)",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"num_nonzero_vars\"`. ",
+          "default": "num_nonzero_vars"
+        },
+        "output_obs_total_counts_vars": {
+          "type": "string",
+          "description": "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"total_counts\"`. ",
+          "default": "total_counts"
+        },
+        "output_var_num_nonzero_obs": {
+          "type": "string",
+          "description": "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes)",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"num_nonzero_obs\"`. ",
+          "default": "num_nonzero_obs"
+        },
+        "output_var_total_counts_obs": {
+          "type": "string",
+          "description": "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"total_counts\"`. ",
+          "default": "total_counts"
+        },
+        "output_var_obs_mean": {
+          "type": "string",
+          "description": "Name of the column in .obs providing the mean of the values in each row.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"obs_mean\"`. ",
+          "default": "obs_mean"
+        },
+        "output_var_pct_dropout": {
+          "type": "string",
+          "description": "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"pct_dropout\"`. ",
+          "default": "pct_dropout"
+        }
+      }
+    },
+    "clr arguments": {
+      "title": "CLR arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "clr_axis": {
+          "type": "integer",
+          "description": "Axis across which CLR is performed.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `0`. ",
+          "default": 0
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/qc metrics calculation options"
+    },
+    {
+      "$ref": "#/$defs/clr arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_multisample/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/.config.vsh.yaml
new file mode 100644
index 0000000..ce978ca
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/.config.vsh.yaml
@@ -0,0 +1,306 @@
+name: "prot_singlesample"
+namespace: "workflows/prot"
+version: "v4.0.4"
+authors:
+- name: "Dries De Maeyer"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Input"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the sample."
+    info: null
+    example:
+    - "foo"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    description: "Path to the sample."
+    info: null
+    example:
+    - "dataset.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Input layer to start from. By default, .X will be used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Output"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Filtering options"
+  arguments:
+  - type: "integer"
+    name: "--min_counts"
+    description: "Minimum number of counts captured per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_counts"
+    description: "Maximum number of counts captured per cell."
+    info: null
+    example:
+    - 5000000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_proteins_per_cell"
+    description: "Minimum of non-zero values per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_proteins_per_cell"
+    description: "Maximum of non-zero values per cell."
+    info: null
+    example:
+    - 1500000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_cells_per_protein"
+    description: "Minimum of non-zero values per gene."
+    info: null
+    example:
+    - 3
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Processing unimodal single-sample CITE-seq data."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info:
+  image: "/images/concepts/fig_workflow_multiomics_adt_singlesample.svg"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "filter/filter_with_counts"
+  repository:
+    type: "local"
+- name: "filter/do_filter"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/prot/prot_singlesample/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/prot/prot_singlesample"
+  executable: "target/nextflow/workflows/prot/prot_singlesample/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/filter/filter_with_counts"
+  - "target/nextflow/filter/do_filter"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/main.nf
new file mode 100644
index 0000000..cdaa8fa
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/main.nf
@@ -0,0 +1,3622 @@
+// prot_singlesample v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries De Maeyer (author)
+//  * Robrecht Cannoodt (author, maintainer)
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "prot_singlesample",
+  "namespace" : "workflows/prot",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Input",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the sample.",
+          "example" : [
+            "foo"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Path to the sample.",
+          "example" : [
+            "dataset.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Input layer to start from. By default, .X will be used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Output",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Filtering options",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--min_counts",
+          "description" : "Minimum number of counts captured per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_counts",
+          "description" : "Maximum number of counts captured per cell.",
+          "example" : [
+            5000000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_proteins_per_cell",
+          "description" : "Minimum of non-zero values per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_proteins_per_cell",
+          "description" : "Maximum of non-zero values per cell.",
+          "example" : [
+            1500000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_cells_per_protein",
+          "description" : "Minimum of non-zero values per gene.",
+          "example" : [
+            3
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Processing unimodal single-sample CITE-seq data.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "info" : {
+    "image" : "/images/concepts/fig_workflow_multiomics_adt_singlesample.svg"
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "filter/filter_with_counts",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "filter/do_filter",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/prot/prot_singlesample/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/prot/prot_singlesample",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { filter_with_counts } from "${meta.resources_dir}/../../../../nextflow/filter/filter_with_counts/main.nf"
+include { do_filter } from "${meta.resources_dir}/../../../../nextflow/filter/do_filter/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+  output_ch = input_ch
+    | map {id, state ->
+      def new_state = state + ["workflow_output": state.output]
+      [id, new_state]
+    }
+    // filtering
+    | filter_with_counts.run(
+      key: "prot_filter_with_counts",
+      fromState: { id, state ->
+        def newState = [
+          "input": state.input,
+          "min_counts": state.min_counts,
+          "max_counts": state.max_counts,
+          "min_genes_per_cell": state.min_proteins_per_cell,
+          "max_genes_per_cell": state.max_proteins_per_cell,
+          "min_cells_per_gene": state.min_cells_per_protein,
+          "obs_name_filter": "filter_with_counts",
+          "var_name_filter": "filter_with_counts",
+          "modality": "prot",
+          "layer": state.layer,
+        ]
+        newState
+      },
+      toState: ["input": "output"]
+    )
+    | do_filter.run(
+      key: "prot_do_filter",
+      fromState : { id, state ->
+        // do_filter does not need a layer argument because it filters all layers
+        // from a modality.
+        def newState = [
+          "input": state.input,
+          "obs_filter": "filter_with_counts",
+          "modality": "prot",
+          "var_filter": "filter_with_counts",
+          "output_compression": "gzip",
+          "output": state.workflow_output
+        ]
+        return newState
+      },
+      toState: ["output": "output"],
+    )
+    | setState(["output"])
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/nextflow.config
new file mode 100644
index 0000000..861dee9
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/prot/prot_singlesample'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Processing unimodal single-sample CITE-seq data.'
+  author = 'Dries De Maeyer, Robrecht Cannoodt, Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/nextflow_schema.json
new file mode 100644
index 0000000..988222c
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/nextflow_schema.json
@@ -0,0 +1,104 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "prot_singlesample",
+  "description": "Processing unimodal single-sample CITE-seq data.",
+  "type": "object",
+  "$defs": {
+    "input": {
+      "title": "Input",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "ID of the sample.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"foo\"`. "
+        },
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the sample.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"dataset.h5mu\"`. "
+        },
+        "layer": {
+          "type": "string",
+          "description": "Input layer to start from",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "output": {
+      "title": "Output",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "filtering options": {
+      "title": "Filtering options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "min_counts": {
+          "type": "integer",
+          "description": "Minimum number of counts captured per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "max_counts": {
+          "type": "integer",
+          "description": "Maximum number of counts captured per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `5000000`. "
+        },
+        "min_proteins_per_cell": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "max_proteins_per_cell": {
+          "type": "integer",
+          "description": "Maximum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `1500000`. "
+        },
+        "min_cells_per_protein": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per gene.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3`. "
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/input"
+    },
+    {
+      "$ref": "#/$defs/output"
+    },
+    {
+      "$ref": "#/$defs/filtering options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/prot/prot_singlesample/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/.config.vsh.yaml
new file mode 100644
index 0000000..3f8034a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/.config.vsh.yaml
@@ -0,0 +1,415 @@
+name: "qc"
+namespace: "workflows/qc"
+version: "v4.0.4"
+authors:
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the sample."
+    info: null
+    example:
+    - "foo"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    alternatives:
+    - "-i"
+    description: "Path to the sample."
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Which modality to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Layer to calculate qc metrics for."
+    info: null
+    example:
+    - "raw_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Mitochondrial & Ribosomal Gene Detection"
+  arguments:
+  - type: "string"
+    name: "--var_gene_names"
+    description: ".var column name to be used to detect mitochondrial/ribosomal genes\
+      \ instead of .var_names (default if not set).\nGene names matching with the\
+      \ regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will\
+      \ be \nidentified as mitochondrial or ribosomal genes, respectively.\n"
+    info: null
+    example:
+    - "gene_symbol"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_mitochondrial_genes"
+    description: "In which .var slot to store a boolean array corresponding the mitochondrial\
+      \ genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_mitochondrial_fraction"
+    description: ".Obs slot to store the fraction of reads found to be mitochondrial.\
+      \ Defaults to 'fraction_' suffixed by the value of --var_name_mitochondrial_genes\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--mitochondrial_gene_regex"
+    description: "Regex string that identifies mitochondrial genes from --var_gene_names.\n\
+      By default will detect human and mouse mitochondrial genes from a gene symbol.\n"
+    info: null
+    default:
+    - "^[mM][tT]-"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_ribosomal_genes"
+    description: "In which .var slot to store a boolean array corresponding the ribosomal\
+      \ genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_ribosomal_fraction"
+    description: "When specified, write the fraction of counts originating from ribosomal\
+      \ genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified\
+      \ name.\nRequires --var_name_ribosomal_genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ribosomal_gene_regex"
+    description: "Regex string that identifies ribosomal genes from --var_gene_names.\n\
+      By default will detect human and mouse ribosomal genes from a gene symbol.\n"
+    info: null
+    default:
+    - "^[Mm]?[Rr][Pp][LlSs]"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "QC metrics calculation options"
+  arguments:
+  - type: "string"
+    name: "--var_qc_metrics"
+    description: "Keys to select a boolean (containing only True or False) column\
+      \ from .var.\nFor each cell, calculate the proportion of total values for genes\
+      \ which are labeled 'True', \ncompared to the total sum of the values for all\
+      \ genes. Defaults to the value from\n--var_name_mitochondrial_genes.\n"
+    info: null
+    example:
+    - "ercc,highly_variable"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+  - type: "integer"
+    name: "--top_n_vars"
+    description: "Number of top vars to be used to calculate cumulative proportions.\n\
+      If not specified, proportions are not calculated. `--top_n_vars 20,50` finds\n\
+      cumulative proportion to the 20th and 50th most expressed vars.\n"
+    info: null
+    default:
+    - 50
+    - 100
+    - 200
+    - 500
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+  - type: "string"
+    name: "--output_obs_num_nonzero_vars"
+    description: "Name of column in .obs describing, for each observation, the number\
+      \ of stored values\n(including explicit zeroes). In other words, the name of\
+      \ the column that counts\nfor each row the number of columns that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_vars"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_total_counts_vars"
+    description: "Name of the column for .obs describing, for each observation (row),\n\
+      the sum of the stored values in the columns.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_num_nonzero_obs"
+    description: "Name of column describing, for each feature, the number of stored\
+      \ values\n(including explicit zeroes). In other words, the name of the column\
+      \ that counts\nfor each column the number of rows that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_obs"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_total_counts_obs"
+    description: "Name of the column in .var describing, for each feature (column),\n\
+      the sum of the stored values in the rows.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_obs_mean"
+    description: "Name of the column in .obs providing the mean of the values in each\
+      \ row.\n"
+    info: null
+    default:
+    - "obs_mean"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_pct_dropout"
+    description: "Name of the column in .obs providing for each feature the percentage\
+      \ of\nobservations the feature does not appear on (i.e. is missing). Same as\
+      \ `--output_var_num_nonzero_obs`\nbut percentage based.\n"
+    info: null
+    default:
+    - "pct_dropout"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "A pipeline to add basic qc statistics to a MuData "
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "concat_test_data"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info:
+  test_dependencies:
+  - name: "qc_test"
+    namespace: "test_workflows/qc"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "metadata/grep_annotation_column"
+  repository:
+    type: "local"
+- name: "qc/calculate_qc_metrics"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/qc/qc/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/qc/qc"
+  executable: "target/nextflow/workflows/qc/qc/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/metadata/grep_annotation_column"
+  - "target/nextflow/qc/calculate_qc_metrics"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/main.nf
new file mode 100644
index 0000000..51a9e71
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/main.nf
@@ -0,0 +1,3766 @@
+// qc v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries Schaumont (author, maintainer)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "qc",
+  "namespace" : "workflows/qc",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the sample.",
+          "example" : [
+            "foo"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "alternatives" : [
+            "-i"
+          ],
+          "description" : "Path to the sample.",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Which modality to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Layer to calculate qc metrics for.",
+          "example" : [
+            "raw_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Mitochondrial & Ribosomal Gene Detection",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_gene_names",
+          "description" : ".var column name to be used to detect mitochondrial/ribosomal genes instead of .var_names (default if not set).\nGene names matching with the regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will be \nidentified as mitochondrial or ribosomal genes, respectively.\n",
+          "example" : [
+            "gene_symbol"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_mitochondrial_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_mitochondrial_fraction",
+          "description" : ".Obs slot to store the fraction of reads found to be mitochondrial. Defaults to 'fraction_' suffixed by the value of --var_name_mitochondrial_genes\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--mitochondrial_gene_regex",
+          "description" : "Regex string that identifies mitochondrial genes from --var_gene_names.\nBy default will detect human and mouse mitochondrial genes from a gene symbol.\n",
+          "default" : [
+            "^[mM][tT]-"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_ribosomal_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_ribosomal_fraction",
+          "description" : "When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified name.\nRequires --var_name_ribosomal_genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--ribosomal_gene_regex",
+          "description" : "Regex string that identifies ribosomal genes from --var_gene_names.\nBy default will detect human and mouse ribosomal genes from a gene symbol.\n",
+          "default" : [
+            "^[Mm]?[Rr][Pp][LlSs]"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "QC metrics calculation options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_qc_metrics",
+          "description" : "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes. Defaults to the value from\n--var_name_mitochondrial_genes.\n",
+          "example" : [
+            "ercc,highly_variable"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        },
+        {
+          "type" : "integer",
+          "name" : "--top_n_vars",
+          "description" : "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated. `--top_n_vars 20,50` finds\ncumulative proportion to the 20th and 50th most expressed vars.\n",
+          "default" : [
+            50,
+            100,
+            200,
+            500
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_num_nonzero_vars",
+          "description" : "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each row the number of columns that contain data.\n",
+          "default" : [
+            "num_nonzero_vars"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_total_counts_vars",
+          "description" : "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_num_nonzero_obs",
+          "description" : "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each column the number of rows that contain data.\n",
+          "default" : [
+            "num_nonzero_obs"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_total_counts_obs",
+          "description" : "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_obs_mean",
+          "description" : "Name of the column in .obs providing the mean of the values in each row.\n",
+          "default" : [
+            "obs_mean"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_pct_dropout",
+          "description" : "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e. is missing). Same as `--output_var_num_nonzero_obs`\nbut percentage based.\n",
+          "default" : [
+            "pct_dropout"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "A pipeline to add basic qc statistics to a MuData ",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "info" : {
+    "test_dependencies" : [
+      {
+        "name" : "qc_test",
+        "namespace" : "test_workflows/qc"
+      }
+    ]
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "metadata/grep_annotation_column",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "qc/calculate_qc_metrics",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/qc/qc/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/qc/qc",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { grep_annotation_column } from "${meta.resources_dir}/../../../../nextflow/metadata/grep_annotation_column/main.nf"
+include { calculate_qc_metrics } from "${meta.resources_dir}/../../../../nextflow/qc/calculate_qc_metrics/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+    output_ch = input_ch
+    // Avoid conflict between output from component and output for this workflow
+    | map {id, state -> 
+      assert state.output, "Output must be defined"
+      def new_state = state + ["workflow_output": state.output]
+      [id, new_state]
+    }
+    // Add default for var_qc_metrics component
+    | map {id, state ->
+      def var_qc_default = []
+      // Remove the var_qc_metric argument from the state if its value is null (not specified)
+      def new_state = state.findAll { it.key != "var_qc_metrics" || it.value == null }
+      if (state.var_name_mitochondrial_genes) {
+        var_qc_default.add(state.var_name_mitochondrial_genes)
+      }
+      if (state.var_name_ribosomal_genes) {
+        var_qc_default.add(state.var_name_ribosomal_genes)
+      }
+      // Get the new state, but make sure to overwrite var_qc_metrics if the user has set it.
+      new_state = ["var_qc_metrics": var_qc_default.join(";")] + new_state
+      [id, new_state]
+    }
+
+    | grep_annotation_column.run(
+      key: "grep_mitochondrial_genes",
+      runIf: { id, state ->
+        state.var_name_mitochondrial_genes
+      },
+      fromState: { id, state ->
+        def stateMapping = [
+          "input": state.input,
+          "modality": state.modality,
+          "input_column": state.var_gene_names,
+          "matrix": "var",
+          "output_match_column": state.var_name_mitochondrial_genes,
+          "regex_pattern": state.mitochondrial_gene_regex,
+          "input_layer": state.layer,
+        ]
+        stateMapping.output_fraction_column = state.obs_name_mitochondrial_fraction ? state.obs_name_mitochondrial_fraction: "fraction_$state.var_name_mitochondrial_genes"
+        return stateMapping
+      },
+      toState: ["input": "output"]
+    )
+
+    | grep_annotation_column.run(
+      key: "grep_ribosomal_genes",
+      runIf: { id, state ->
+        state.var_name_ribosomal_genes
+      },
+      fromState: { id, state ->
+        def stateMapping = [
+          "input": state.input,
+          "modality": state.modality,
+          "input_column": state.var_gene_names,
+          "matrix": "var",
+          "output_match_column": state.var_name_ribosomal_genes,
+          "regex_pattern": state.ribosomal_gene_regex,
+          "input_layer": state.layer,
+        ]
+        stateMapping.output_fraction_column = state.obs_name_ribosomal_fraction ? state.obs_name_ribosomal_fraction: "fraction_$state.var_name_ribosomal_genes"
+        return stateMapping
+      },
+      toState: ["input": "output"]
+    )
+
+    | calculate_qc_metrics.run(
+      fromState: { id, state ->
+        def newState = [
+          "input": state.input,
+          "modality": state.modality,
+          "layer": state.layer,
+          // TODO: remove this workaround when Viash issue is resolved:
+          //       'top_n_vars': list(map(int, r''.split(';'))),
+          //     ValueError: invalid literal for int() with base 10: ''
+          // See https://github.com/viash-io/viash/issues/619
+          "top_n_vars": state.top_n_vars ? state.top_n_vars : null,
+          "var_qc_metrics_fill_na_value": state.var_qc_metrics_fill_na_value,
+          "output_obs_num_nonzero_vars": state.output_obs_num_nonzero_vars,
+          "output_obs_total_counts_vars": state.output_obs_total_counts_vars,
+          "output_var_num_nonzero_obs": state.output_var_num_nonzero_obs,
+          "output_var_total_counts_obs": state.output_var_total_counts_obs,
+          "output_var_obs_mean": state.output_var_obs_mean,
+          "output_var_pct_dropout": state.output_var_pct_dropout,
+          "output": state.workflow_output,
+          "compression": "gzip"
+        ]
+        if (state.var_qc_metrics) {
+          newState += ["var_qc_metrics": state.var_qc_metrics]
+        }
+      return newState
+      },
+      toState: ["output": "output"]
+    )
+    | setState(["output"]) 
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/nextflow.config
new file mode 100644
index 0000000..7735547
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/qc/qc'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'A pipeline to add basic qc statistics to a MuData '
+  author = 'Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/nextflow_schema.json
new file mode 100644
index 0000000..82142b4
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/nextflow_schema.json
@@ -0,0 +1,190 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "qc",
+  "description": "A pipeline to add basic qc statistics to a MuData ",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "ID of the sample.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"foo\"`. "
+        },
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the sample.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Which modality to process.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "layer": {
+          "type": "string",
+          "description": "Layer to calculate qc metrics for.",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"raw_counts\"`. "
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "mitochondrial & ribosomal gene detection": {
+      "title": "Mitochondrial & Ribosomal Gene Detection",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_gene_names": {
+          "type": "string",
+          "description": ".var column name to be used to detect mitochondrial/ribosomal genes instead of .var_names (default if not set).\nGene names matching with the regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will be \nidentified as mitochondrial or ribosomal genes, respectively.\n",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gene_symbol\"`. "
+        },
+        "var_name_mitochondrial_genes": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_name_mitochondrial_fraction": {
+          "type": "string",
+          "description": ".Obs slot to store the fraction of reads found to be mitochondrial",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "mitochondrial_gene_regex": {
+          "type": "string",
+          "description": "Regex string that identifies mitochondrial genes from --var_gene_names.\nBy default will detect human and mouse mitochondrial genes from a gene symbol.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"^[mM][tT]-\"`. ",
+          "default": "^[mM][tT]-"
+        },
+        "var_name_ribosomal_genes": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_name_ribosomal_fraction": {
+          "type": "string",
+          "description": "When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified name.\nRequires --var_name_ribosomal_genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "ribosomal_gene_regex": {
+          "type": "string",
+          "description": "Regex string that identifies ribosomal genes from --var_gene_names.\nBy default will detect human and mouse ribosomal genes from a gene symbol.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"^[Mm]?[Rr][Pp][LlSs]\"`. ",
+          "default": "^[Mm]?[Rr][Pp][LlSs]"
+        }
+      }
+    },
+    "qc metrics calculation options": {
+      "title": "QC metrics calculation options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_qc_metrics": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"ercc,highly_variable\"]`. "
+        },
+        "top_n_vars": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated",
+          "help_text": "Type: `integer`, multiple: `True`, default: `[50,100,200,500]`. ",
+          "default": [
+            50,
+            100,
+            200,
+            500
+          ]
+        },
+        "output_obs_num_nonzero_vars": {
+          "type": "string",
+          "description": "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes)",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"num_nonzero_vars\"`. ",
+          "default": "num_nonzero_vars"
+        },
+        "output_obs_total_counts_vars": {
+          "type": "string",
+          "description": "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"total_counts\"`. ",
+          "default": "total_counts"
+        },
+        "output_var_num_nonzero_obs": {
+          "type": "string",
+          "description": "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes)",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"num_nonzero_obs\"`. ",
+          "default": "num_nonzero_obs"
+        },
+        "output_var_total_counts_obs": {
+          "type": "string",
+          "description": "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"total_counts\"`. ",
+          "default": "total_counts"
+        },
+        "output_var_obs_mean": {
+          "type": "string",
+          "description": "Name of the column in .obs providing the mean of the values in each row.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"obs_mean\"`. ",
+          "default": "obs_mean"
+        },
+        "output_var_pct_dropout": {
+          "type": "string",
+          "description": "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"pct_dropout\"`. ",
+          "default": "pct_dropout"
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/mitochondrial & ribosomal gene detection"
+    },
+    {
+      "$ref": "#/$defs/qc metrics calculation options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/qc/qc/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/.config.vsh.yaml
new file mode 100644
index 0000000..7029fed
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/.config.vsh.yaml
@@ -0,0 +1,476 @@
+name: "rna_multisample"
+namespace: "workflows/rna"
+version: "v4.0.4"
+authors:
+- name: "Dries De Maeyer"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the concatenated file"
+    info: null
+    example:
+    - "concatenated"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    description: "Path to the samples."
+    info: null
+    example:
+    - "dataset.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Modality to process."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Input layer to use. If not specified, .X is used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Output"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Filtering highly variable features"
+  arguments:
+  - type: "string"
+    name: "--highly_variable_features_var_output"
+    alternatives:
+    - "--filter_with_hvg_var_output"
+    description: "In which .var slot to store a boolean array corresponding to the\
+      \ highly variable features."
+    info: null
+    default:
+    - "filter_with_hvg"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--highly_variable_features_obs_batch_key"
+    alternatives:
+    - "--filter_with_hvg_obs_batch_key"
+    description: "If specified, highly-variable features are selected within each\
+      \ batch separately and merged. This simple \nprocess avoids the selection of\
+      \ batch-specific features and acts as a lightweight batch correction method.\
+      \ \nFor all flavors, featues are first sorted by how many batches they are highly\
+      \ variable. For dispersion-based flavors \nties are broken by normalized dispersion.\
+      \ If flavor = 'seurat_v3', ties are broken by the median (across\nbatches) rank\
+      \ based on within-batch normalized variance.\n"
+    info: null
+    default:
+    - "sample_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--highly_variable_features_flavor"
+    alternatives:
+    - "--filter_with_hvg_flavor"
+    description: "Choose the flavor for identifying highly variable features. For\
+      \ the dispersion based methods\nin their default workflows, Seurat passes the\
+      \ cutoffs whereas Cell Ranger passes n_top_features.\n"
+    info: null
+    default:
+    - "seurat"
+    required: false
+    choices:
+    - "seurat"
+    - "cell_ranger"
+    - "seurat_v3"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--highly_variable_features_n_top_features"
+    alternatives:
+    - "--filter_with_hvg_n_top_genes"
+    description: "Number of highly-variable features to keep. Mandatory if filter_with_hvg_flavor\
+      \ is set to 'seurat_v3'."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "QC metrics calculation options"
+  arguments:
+  - type: "string"
+    name: "--var_qc_metrics"
+    description: "Keys to select a boolean (containing only True or False) column\
+      \ from .var.\nFor each cell, calculate the proportion of total values for genes\
+      \ which are labeled 'True', \ncompared to the total sum of the values for all\
+      \ genes.\n"
+    info: null
+    example:
+    - "ercc,highly_variable"
+    default:
+    - "filter_with_hvg"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+  - type: "integer"
+    name: "--top_n_vars"
+    description: "Number of top vars to be used to calculate cumulative proportions.\n\
+      If not specified, proportions are not calculated. `--top_n_vars 20,50` finds\n\
+      cumulative proportion to the 20th and 50th most expressed vars.\n"
+    info: null
+    default:
+    - 50
+    - 100
+    - 200
+    - 500
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+  - type: "string"
+    name: "--output_obs_num_nonzero_vars"
+    description: "Name of column in .obs describing, for each observation, the number\
+      \ of stored values\n(including explicit zeroes). In other words, the name of\
+      \ the column that counts\nfor each row the number of columns that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_vars"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_obs_total_counts_vars"
+    description: "Name of the column for .obs describing, for each observation (row),\n\
+      the sum of the stored values in the columns.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_num_nonzero_obs"
+    description: "Name of column describing, for each feature, the number of stored\
+      \ values\n(including explicit zeroes). In other words, the name of the column\
+      \ that counts\nfor each column the number of rows that contain data.\n"
+    info: null
+    default:
+    - "num_nonzero_obs"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_total_counts_obs"
+    description: "Name of the column in .var describing, for each feature (column),\n\
+      the sum of the stored values in the rows.\n"
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_obs_mean"
+    description: "Name of the column in .obs providing the mean of the values in each\
+      \ row.\n"
+    info: null
+    default:
+    - "obs_mean"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--output_var_pct_dropout"
+    description: "Name of the column in .obs providing for each feature the percentage\
+      \ of\nobservations the feature does not appear on (i.e. is missing). Same as\
+      \ `--num_nonzero_obs`\nbut percentage based.\n"
+    info: null
+    default:
+    - "pct_dropout"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "RNA Scaling options"
+  description: "Options for enabling scaling of the log-normalized data to unit variance\
+    \ and zero mean.\nThe scaled data will be output a different layer and representation\
+    \ with reduced dimensions\nwill be created and stored in addition to the non-scaled\
+    \ data.\n"
+  arguments:
+  - type: "boolean_true"
+    name: "--enable_scaling"
+    description: "Enable scaling for the RNA modality."
+    info: null
+    direction: "input"
+  - type: "string"
+    name: "--scaling_output_layer"
+    description: "Output layer where the scaled log-normalized data will be stored."
+    info: null
+    default:
+    - "scaled"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--scaling_max_value"
+    description: "Clip (truncate) data to this value after scaling. If not specified,\
+      \ do not clip."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_false"
+    name: "--scaling_zero_center"
+    description: "If set, omit zero-centering variables, which allows to handle sparse\
+      \ input efficiently.\""
+    info: null
+    direction: "input"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Processing unimodal multi-sample RNA transcriptomics data."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "concat_test_data"
+info:
+  image: "/images/concepts/fig_workflow_multiomics_rna_multisample.svg"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "feature_annotation/highly_variable_features_scanpy"
+  repository:
+    type: "local"
+- name: "workflows/qc/qc"
+  alias: "rna_qc"
+  repository:
+    type: "local"
+- name: "metadata/add_id"
+  repository:
+    type: "local"
+- name: "transform/scale"
+  repository:
+    type: "local"
+- name: "workflows/rna/log_normalize"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/rna/rna_multisample/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/rna/rna_multisample"
+  executable: "target/nextflow/workflows/rna/rna_multisample/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/feature_annotation/highly_variable_features_scanpy"
+  - "target/nextflow/workflows/qc/qc"
+  - "target/nextflow/metadata/add_id"
+  - "target/nextflow/transform/scale"
+  - "target/_private/nextflow/workflows/rna/log_normalize"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/main.nf
new file mode 100644
index 0000000..f12a144
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/main.nf
@@ -0,0 +1,3829 @@
+// rna_multisample v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries De Maeyer (author)
+//  * Robrecht Cannoodt (author, maintainer)
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "rna_multisample",
+  "namespace" : "workflows/rna",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the concatenated file",
+          "example" : [
+            "concatenated"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Path to the samples.",
+          "example" : [
+            "dataset.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Modality to process.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Input layer to use. If not specified, .X is used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Output",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Filtering highly variable features",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--highly_variable_features_var_output",
+          "alternatives" : [
+            "--filter_with_hvg_var_output"
+          ],
+          "description" : "In which .var slot to store a boolean array corresponding to the highly variable features.",
+          "default" : [
+            "filter_with_hvg"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--highly_variable_features_obs_batch_key",
+          "alternatives" : [
+            "--filter_with_hvg_obs_batch_key"
+          ],
+          "description" : "If specified, highly-variable features are selected within each batch separately and merged. This simple \nprocess avoids the selection of batch-specific features and acts as a lightweight batch correction method. \nFor all flavors, featues are first sorted by how many batches they are highly variable. For dispersion-based flavors \nties are broken by normalized dispersion. If flavor = 'seurat_v3', ties are broken by the median (across\nbatches) rank based on within-batch normalized variance.\n",
+          "default" : [
+            "sample_id"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--highly_variable_features_flavor",
+          "alternatives" : [
+            "--filter_with_hvg_flavor"
+          ],
+          "description" : "Choose the flavor for identifying highly variable features. For the dispersion based methods\nin their default workflows, Seurat passes the cutoffs whereas Cell Ranger passes n_top_features.\n",
+          "default" : [
+            "seurat"
+          ],
+          "required" : false,
+          "choices" : [
+            "seurat",
+            "cell_ranger",
+            "seurat_v3"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--highly_variable_features_n_top_features",
+          "alternatives" : [
+            "--filter_with_hvg_n_top_genes"
+          ],
+          "description" : "Number of highly-variable features to keep. Mandatory if filter_with_hvg_flavor is set to 'seurat_v3'.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "QC metrics calculation options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_qc_metrics",
+          "description" : "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes.\n",
+          "example" : [
+            "ercc,highly_variable"
+          ],
+          "default" : [
+            "filter_with_hvg"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        },
+        {
+          "type" : "integer",
+          "name" : "--top_n_vars",
+          "description" : "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated. `--top_n_vars 20,50` finds\ncumulative proportion to the 20th and 50th most expressed vars.\n",
+          "default" : [
+            50,
+            100,
+            200,
+            500
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_num_nonzero_vars",
+          "description" : "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each row the number of columns that contain data.\n",
+          "default" : [
+            "num_nonzero_vars"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_obs_total_counts_vars",
+          "description" : "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_num_nonzero_obs",
+          "description" : "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes). In other words, the name of the column that counts\nfor each column the number of rows that contain data.\n",
+          "default" : [
+            "num_nonzero_obs"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_total_counts_obs",
+          "description" : "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_obs_mean",
+          "description" : "Name of the column in .obs providing the mean of the values in each row.\n",
+          "default" : [
+            "obs_mean"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--output_var_pct_dropout",
+          "description" : "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e. is missing). Same as `--num_nonzero_obs`\nbut percentage based.\n",
+          "default" : [
+            "pct_dropout"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "RNA Scaling options",
+      "description" : "Options for enabling scaling of the log-normalized data to unit variance and zero mean.\nThe scaled data will be output a different layer and representation with reduced dimensions\nwill be created and stored in addition to the non-scaled data.\n",
+      "arguments" : [
+        {
+          "type" : "boolean_true",
+          "name" : "--enable_scaling",
+          "description" : "Enable scaling for the RNA modality.",
+          "direction" : "input"
+        },
+        {
+          "type" : "string",
+          "name" : "--scaling_output_layer",
+          "description" : "Output layer where the scaled log-normalized data will be stored.",
+          "default" : [
+            "scaled"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--scaling_max_value",
+          "description" : "Clip (truncate) data to this value after scaling. If not specified, do not clip.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_false",
+          "name" : "--scaling_zero_center",
+          "description" : "If set, omit zero-centering variables, which allows to handle sparse input efficiently.\\"",
+          "direction" : "input"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Processing unimodal multi-sample RNA transcriptomics data.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/concat_test_data"
+    }
+  ],
+  "info" : {
+    "image" : "/images/concepts/fig_workflow_multiomics_rna_multisample.svg"
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "feature_annotation/highly_variable_features_scanpy",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/qc/qc",
+      "alias" : "rna_qc",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "metadata/add_id",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "transform/scale",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/rna/log_normalize",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/rna/rna_multisample/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/rna/rna_multisample",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { highly_variable_features_scanpy } from "${meta.resources_dir}/../../../../nextflow/feature_annotation/highly_variable_features_scanpy/main.nf"
+include { qc as rna_qc_viashalias } from "${meta.resources_dir}/../../../../nextflow/workflows/qc/qc/main.nf"
+rna_qc = rna_qc_viashalias.run(key: "rna_qc")
+include { add_id } from "${meta.resources_dir}/../../../../nextflow/metadata/add_id/main.nf"
+include { scale } from "${meta.resources_dir}/../../../../nextflow/transform/scale/main.nf"
+include { log_normalize } from "${meta.resources_dir}/../../../../_private/nextflow/workflows/rna/log_normalize/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+  output_ch = input_ch
+    | map {id, state ->
+      def new_state = state + ["workflow_output": state.output]
+      [id, new_state]
+    }
+    | log_normalize.run(
+      args: ["output_layer": "log_normalized"],
+      fromState: [
+        "input": "input",
+        "layer": "layer",
+        "modality": "modality"
+      ],
+      toState: [
+        "input": "output"
+      ]
+    )
+    | scale.run(
+      runIf: {id, state -> state.enable_scaling},
+      fromState: {id, state -> 
+        [
+          "input": state.input,
+          "modality": state.modality,
+          "input_layer": "log_normalized",
+          "output_layer": state.scaling_output_layer,
+          "max_value": state.scaling_max_value,
+          "zero_center": state.scaling_zero_center,
+        ]
+      },
+      toState: ["input": "output"],
+    )
+    | highly_variable_features_scanpy.run(
+      fromState: {id, state ->
+        [
+          "input": state.input,
+          "layer": "log_normalized",
+          "modality": state.modality,
+          "var_name_filter": state.highly_variable_features_var_output,
+          "n_top_features": state.highly_variable_features_n_top_features,
+          "flavor": state.highly_variable_features_flavor,
+          "obs_batch_key": state.highly_variable_features_obs_batch_key
+        ]
+      },
+      toState: ["input": "output"],
+    )
+    | rna_qc.run(
+      // TODO: remove when viash 0.8.3 is released
+      key: "rna_qc",
+      fromState: {id, state ->
+        [
+          "id": id,
+          "input": state.input,
+          "output": state.workflow_output,
+          "layer": state.layer, // Use the non-transformed layer
+          "output_compression": "gzip",
+          "modality": state.modality,
+          "var_qc_metrics": state.var_qc_metrics,
+          "top_n_vars": state.top_n_vars,
+          "output_obs_num_nonzero_vars": state.output_obs_num_nonzero_vars,
+          "output_obs_total_counts_vars": state.output_obs_total_counts_vars,
+          "output_var_num_nonzero_obs": state.output_var_num_nonzero_obs,
+          "output_var_total_counts_obs": state.output_var_total_counts_obs,
+          "output_var_obs_mean": state.output_var_obs_mean,
+          "output_var_pct_dropout": state.output_var_pct_dropout
+        ]
+      },
+    )
+    | setState(["output"])
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/nextflow.config
new file mode 100644
index 0000000..eb441bb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/rna/rna_multisample'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Processing unimodal multi-sample RNA transcriptomics data.'
+  author = 'Dries De Maeyer, Robrecht Cannoodt, Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/nextflow_schema.json
new file mode 100644
index 0000000..8d790f5
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/nextflow_schema.json
@@ -0,0 +1,217 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "rna_multisample",
+  "description": "Processing unimodal multi-sample RNA transcriptomics data.",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "ID of the concatenated file",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"concatenated\"`. "
+        },
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the samples.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"dataset.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Modality to process.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "layer": {
+          "type": "string",
+          "description": "Input layer to use",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "output": {
+      "title": "Output",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "filtering highly variable features": {
+      "title": "Filtering highly variable features",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "highly_variable_features_var_output": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding to the highly variable features.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"filter_with_hvg\"`. ",
+          "default": "filter_with_hvg"
+        },
+        "highly_variable_features_obs_batch_key": {
+          "type": "string",
+          "description": "If specified, highly-variable features are selected within each batch separately and merged",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"sample_id\"`. ",
+          "default": "sample_id"
+        },
+        "highly_variable_features_flavor": {
+          "type": "string",
+          "description": "Choose the flavor for identifying highly variable features",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"seurat\"`, choices: ``seurat`, `cell_ranger`, `seurat_v3``. ",
+          "enum": [
+            "seurat",
+            "cell_ranger",
+            "seurat_v3"
+          ],
+          "default": "seurat"
+        },
+        "highly_variable_features_n_top_features": {
+          "type": "integer",
+          "description": "Number of highly-variable features to keep",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        }
+      }
+    },
+    "qc metrics calculation options": {
+      "title": "QC metrics calculation options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_qc_metrics": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes.\n",
+          "help_text": "Type: `string`, multiple: `True`, default: `[\"filter_with_hvg\"]`, example: `[\"ercc,highly_variable\"]`. ",
+          "default": [
+            "filter_with_hvg"
+          ]
+        },
+        "top_n_vars": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Number of top vars to be used to calculate cumulative proportions.\nIf not specified, proportions are not calculated",
+          "help_text": "Type: `integer`, multiple: `True`, default: `[50,100,200,500]`. ",
+          "default": [
+            50,
+            100,
+            200,
+            500
+          ]
+        },
+        "output_obs_num_nonzero_vars": {
+          "type": "string",
+          "description": "Name of column in .obs describing, for each observation, the number of stored values\n(including explicit zeroes)",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"num_nonzero_vars\"`. ",
+          "default": "num_nonzero_vars"
+        },
+        "output_obs_total_counts_vars": {
+          "type": "string",
+          "description": "Name of the column for .obs describing, for each observation (row),\nthe sum of the stored values in the columns.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"total_counts\"`. ",
+          "default": "total_counts"
+        },
+        "output_var_num_nonzero_obs": {
+          "type": "string",
+          "description": "Name of column describing, for each feature, the number of stored values\n(including explicit zeroes)",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"num_nonzero_obs\"`. ",
+          "default": "num_nonzero_obs"
+        },
+        "output_var_total_counts_obs": {
+          "type": "string",
+          "description": "Name of the column in .var describing, for each feature (column),\nthe sum of the stored values in the rows.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"total_counts\"`. ",
+          "default": "total_counts"
+        },
+        "output_var_obs_mean": {
+          "type": "string",
+          "description": "Name of the column in .obs providing the mean of the values in each row.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"obs_mean\"`. ",
+          "default": "obs_mean"
+        },
+        "output_var_pct_dropout": {
+          "type": "string",
+          "description": "Name of the column in .obs providing for each feature the percentage of\nobservations the feature does not appear on (i.e",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"pct_dropout\"`. ",
+          "default": "pct_dropout"
+        }
+      }
+    },
+    "rna scaling options": {
+      "title": "RNA Scaling options",
+      "type": "object",
+      "description": "Options for enabling scaling of the log-normalized data to unit variance and zero mean.\nThe scaled data will be output a different layer and representation with reduced dimensions\nwill be created and stored in addition to the non-scaled data.\n",
+      "properties": {
+        "enable_scaling": {
+          "type": "boolean",
+          "description": "Enable scaling for the RNA modality.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "scaling_output_layer": {
+          "type": "string",
+          "description": "Output layer where the scaled log-normalized data will be stored.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"scaled\"`. ",
+          "default": "scaled"
+        },
+        "scaling_max_value": {
+          "type": "number",
+          "description": "Clip (truncate) data to this value after scaling",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "scaling_zero_center": {
+          "type": "boolean",
+          "description": "If set, omit zero-centering variables, which allows to handle sparse input efficiently.\"",
+          "help_text": "Type: `boolean_false`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/output"
+    },
+    {
+      "$ref": "#/$defs/filtering highly variable features"
+    },
+    {
+      "$ref": "#/$defs/qc metrics calculation options"
+    },
+    {
+      "$ref": "#/$defs/rna scaling options"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_multisample/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/.config.vsh.yaml
new file mode 100644
index 0000000..bf58927
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/.config.vsh.yaml
@@ -0,0 +1,450 @@
+name: "rna_singlesample"
+namespace: "workflows/rna"
+version: "v4.0.4"
+authors:
+- name: "Dries De Maeyer"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "ddemaeyer@gmail.com"
+      github: "ddemaeyer"
+      linkedin: "dries-de-maeyer-b46a814"
+    organizations:
+    - name: "Janssen Pharmaceuticals"
+      href: "https://www.janssen.com"
+      role: "Principal Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Dries Schaumont"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dries@data-intuitive.com"
+      github: "DriesSchaumont"
+      orcid: "0000-0002-4389-0440"
+      linkedin: "dries-schaumont"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+argument_groups:
+- name: "Input"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the sample."
+    info: null
+    example:
+    - "foo"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    description: "Path to the sample."
+    info: null
+    example:
+    - "dataset.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--layer"
+    description: "Input layer to start from. By default, .X will be used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Output"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "Destination path to the output."
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Filtering options"
+  arguments:
+  - type: "integer"
+    name: "--min_counts"
+    description: "Minimum number of counts captured per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_counts"
+    description: "Maximum number of counts captured per cell."
+    info: null
+    example:
+    - 5000000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_genes_per_cell"
+    description: "Minimum of non-zero values per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_genes_per_cell"
+    description: "Maximum of non-zero values per cell."
+    info: null
+    example:
+    - 1500000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_cells_per_gene"
+    description: "Minimum of non-zero values per gene."
+    info: null
+    example:
+    - 3
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--min_fraction_mito"
+    description: "Minimum fraction of UMIs that are mitochondrial. Requires --obs_name_mitochondrial_fraction."
+    info: null
+    example:
+    - 0.0
+    required: false
+    min: 0.0
+    max: 1.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--max_fraction_mito"
+    description: "Maximum fraction of UMIs that are mitochondrial. Requires --obs_name_mitochondrial_fraction.\n"
+    info: null
+    example:
+    - 0.2
+    required: false
+    min: 0.0
+    max: 1.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--min_fraction_ribo"
+    description: "Minimum fraction of UMIs that are ribosomal. Requires --obs_name_ribosomal_fraction."
+    info: null
+    example:
+    - 0.0
+    required: false
+    min: 0.0
+    max: 1.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--max_fraction_ribo"
+    description: "Maximum fraction of UMIs that are ribosomal. Requires --obs_name_ribosomal_fraction.\n"
+    info: null
+    example:
+    - 0.2
+    required: false
+    min: 0.0
+    max: 1.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--skip_scrublet_doublet_detection"
+    description: "Skip the scrublet doublet detection step."
+    info: null
+    direction: "input"
+- name: "Mitochondrial & Ribosomal Gene Detection"
+  arguments:
+  - type: "string"
+    name: "--var_gene_names"
+    description: ".var column name to be used to detect mitochondrial/ribosomal genes\
+      \ instead of .var_names (default if not set).\nGene names matching with the\
+      \ regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will\
+      \ be \nidentified as mitochondrial or ribosomal genes, respectively.  \n"
+    info: null
+    example:
+    - "gene_symbol"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_mitochondrial_genes"
+    description: "In which .var slot to store a boolean array corresponding the mitochondrial\
+      \ genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_mitochondrial_fraction"
+    description: "When specified, write the fraction of counts originating from mitochondrial\
+      \ genes \n(based on --mitochondrial_gene_regex) to an .obs column with the specified\
+      \ name.\nRequires --var_name_mitochondrial_genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--mitochondrial_gene_regex"
+    description: "Regex string that identifies mitochondrial genes from --var_gene_names.\n\
+      By default will detect human and mouse mitochondrial genes from a gene symbol.\n"
+    info: null
+    default:
+    - "^[mM][tT]-"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_ribosomal_genes"
+    description: "In which .var slot to store a boolean array corresponding the ribosomal\
+      \ genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_ribosomal_fraction"
+    description: "When specified, write the fraction of counts originating from ribosomal\
+      \ genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified\
+      \ name.\nRequires --var_name_ribosomal_genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ribosomal_gene_regex"
+    description: "Regex string that identifies ribosomal genes from --var_gene_names.\n\
+      By default will detect human and mouse ribosomal genes from a gene symbol.\n"
+    info: null
+    default:
+    - "^[Mm]?[Rr][Pp][LlSs]"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "utils"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Processing unimodal single-sample RNA transcriptomics data."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf2"
+- type: "file"
+  path: "pbmc_1k_protein_v3"
+info:
+  image: "/images/concepts/fig_workflow_multiomics_rna_singlesample.svg"
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+dependencies:
+- name: "filter/filter_with_counts"
+  repository:
+    type: "local"
+- name: "filter/filter_with_scrublet"
+  repository:
+    type: "local"
+- name: "filter/do_filter"
+  repository:
+    type: "local"
+- name: "filter/delimit_fraction"
+  repository:
+    type: "local"
+- name: "workflows/qc/qc"
+  repository:
+    type: "local"
+license: "MIT"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/rna/rna_singlesample/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/rna/rna_singlesample"
+  executable: "target/nextflow/workflows/rna/rna_singlesample/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "fb7dc76676aa63d06ae1421bbdd6312ad4f67312"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline"
+  dependencies:
+  - "target/nextflow/filter/filter_with_counts"
+  - "target/nextflow/filter/filter_with_scrublet"
+  - "target/nextflow/filter/do_filter"
+  - "target/nextflow/filter/delimit_fraction"
+  - "target/nextflow/workflows/qc/qc"
+package_config:
+  name: "openpipeline"
+  version: "v4.0.4"
+  summary: "Best-practice workflows for single-cell multi-omics analyses.\n"
+  description: "OpenPipelines are extensible single cell analysis pipelines for reproducible\
+    \ and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\
+    \nIn terms of workflows, the following has been made available, but keep in mind\
+    \ that\nindividual tools and functionality can be executed as standalone components\
+    \ as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n\
+    \  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample\
+    \ processing: cell filtering and doublet detection.\n  * Multisample processing:\
+    \ Count transformation, normalization, QC metric calulations.\n  * Integration:\
+    \ Clustering, integration and batch correction using single and multimodal methods.\n\
+    \  * Downstream analysis workflows\n"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-data"
+      dest: "resources_test"
+    nextflow_labels_ci:
+    - path: "src/workflows/utils/labels_ci.config"
+      description: "Adds the correct memory and CPU labels when running on the Viash\
+        \ Hub CI."
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+  keywords:
+  - "single-cell"
+  - "multimodal"
+  license: "MIT"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline"
+    docker_registry: "ghcr.io"
+    homepage: "https://openpipelines.bio"
+    documentation: "https://openpipelines.bio/fundamentals"
+    issue_tracker: "https://github.com/openpipelines-bio/openpipeline/issues"
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/main.nf b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/main.nf
new file mode 100644
index 0000000..4ef1621
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/main.nf
@@ -0,0 +1,3922 @@
+// rna_singlesample v4.0.4
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dries De Maeyer (author)
+//  * Robrecht Cannoodt (author, maintainer)
+//  * Dries Schaumont (author)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "rna_singlesample",
+  "namespace" : "workflows/rna",
+  "version" : "v4.0.4",
+  "authors" : [
+    {
+      "name" : "Dries De Maeyer",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "ddemaeyer@gmail.com",
+          "github" : "ddemaeyer",
+          "linkedin" : "dries-de-maeyer-b46a814"
+        },
+        "organizations" : [
+          {
+            "name" : "Janssen Pharmaceuticals",
+            "href" : "https://www.janssen.com",
+            "role" : "Principal Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dries Schaumont",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dries@data-intuitive.com",
+          "github" : "DriesSchaumont",
+          "orcid" : "0000-0002-4389-0440",
+          "linkedin" : "dries-schaumont"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Input",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the sample.",
+          "example" : [
+            "foo"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Path to the sample.",
+          "example" : [
+            "dataset.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--layer",
+          "description" : "Input layer to start from. By default, .X will be used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Output",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "Destination path to the output.",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Filtering options",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--min_counts",
+          "description" : "Minimum number of counts captured per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_counts",
+          "description" : "Maximum number of counts captured per cell.",
+          "example" : [
+            5000000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_genes_per_cell",
+          "description" : "Minimum of non-zero values per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_genes_per_cell",
+          "description" : "Maximum of non-zero values per cell.",
+          "example" : [
+            1500000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_cells_per_gene",
+          "description" : "Minimum of non-zero values per gene.",
+          "example" : [
+            3
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--min_fraction_mito",
+          "description" : "Minimum fraction of UMIs that are mitochondrial. Requires --obs_name_mitochondrial_fraction.",
+          "example" : [
+            0.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "max" : 1.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--max_fraction_mito",
+          "description" : "Maximum fraction of UMIs that are mitochondrial. Requires --obs_name_mitochondrial_fraction.\n",
+          "example" : [
+            0.2
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "max" : 1.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--min_fraction_ribo",
+          "description" : "Minimum fraction of UMIs that are ribosomal. Requires --obs_name_ribosomal_fraction.",
+          "example" : [
+            0.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "max" : 1.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--max_fraction_ribo",
+          "description" : "Maximum fraction of UMIs that are ribosomal. Requires --obs_name_ribosomal_fraction.\n",
+          "example" : [
+            0.2
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "max" : 1.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--skip_scrublet_doublet_detection",
+          "description" : "Skip the scrublet doublet detection step.",
+          "direction" : "input"
+        }
+      ]
+    },
+    {
+      "name" : "Mitochondrial & Ribosomal Gene Detection",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_gene_names",
+          "description" : ".var column name to be used to detect mitochondrial/ribosomal genes instead of .var_names (default if not set).\nGene names matching with the regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will be \nidentified as mitochondrial or ribosomal genes, respectively.  \n",
+          "example" : [
+            "gene_symbol"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_mitochondrial_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_mitochondrial_fraction",
+          "description" : "When specified, write the fraction of counts originating from mitochondrial genes \n(based on --mitochondrial_gene_regex) to an .obs column with the specified name.\nRequires --var_name_mitochondrial_genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--mitochondrial_gene_regex",
+          "description" : "Regex string that identifies mitochondrial genes from --var_gene_names.\nBy default will detect human and mouse mitochondrial genes from a gene symbol.\n",
+          "default" : [
+            "^[mM][tT]-"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_ribosomal_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_ribosomal_fraction",
+          "description" : "When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified name.\nRequires --var_name_ribosomal_genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--ribosomal_gene_regex",
+          "description" : "Regex string that identifies ribosomal genes from --var_gene_names.\nBy default will detect human and mouse ribosomal genes from a gene symbol.\n",
+          "default" : [
+            "^[Mm]?[Rr][Pp][LlSs]"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/workflows/utils/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Processing unimodal single-sample RNA transcriptomics data.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf2"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3"
+    }
+  ],
+  "info" : {
+    "image" : "/images/concepts/fig_workflow_multiomics_rna_singlesample.svg"
+  },
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "dependencies" : [
+    {
+      "name" : "filter/filter_with_counts",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "filter/filter_with_scrublet",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "filter/do_filter",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "filter/delimit_fraction",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "workflows/qc/qc",
+      "repository" : {
+        "type" : "local"
+      }
+    }
+  ],
+  "license" : "MIT",
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/rna/rna_singlesample/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/rna/rna_singlesample",
+    "viash_version" : "0.9.4",
+    "git_commit" : "fb7dc76676aa63d06ae1421bbdd6312ad4f67312",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline"
+  },
+  "package_config" : {
+    "name" : "openpipeline",
+    "version" : "v4.0.4",
+    "summary" : "Best-practice workflows for single-cell multi-omics analyses.\n",
+    "description" : "OpenPipelines are extensible single cell analysis pipelines for reproducible and large-scale single cell processing using [Viash](https://viash.io) and [Nextflow](https://www.nextflow.io/).\n\nIn terms of workflows, the following has been made available, but keep in mind that\nindividual tools and functionality can be executed as standalone components as well.\n\n  * Demultiplexing: conversion of raw sequencing data to FASTQ objects.\n  * Ingestion: Read mapping and generating a count matrix.\n  * Single sample processing: cell filtering and doublet detection.\n  * Multisample processing: Count transformation, normalization, QC metric calulations.\n  * Integration: Clustering, integration and batch correction using single and multimodal methods.\n  * Downstream analysis workflows\n",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-data",
+          "dest" : "resources_test"
+        }
+      ],
+      "nextflow_labels_ci" : [
+        {
+          "path" : "src/workflows/utils/labels_ci.config",
+          "description" : "Adds the correct memory and CPU labels when running on the Viash Hub CI."
+        }
+      ]
+    },
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".resources += {path: '/src/workflows/utils/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v4.0.4'"
+    ],
+    "keywords" : [
+      "single-cell",
+      "multimodal"
+    ],
+    "license" : "MIT",
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline",
+      "docker_registry" : "ghcr.io",
+      "homepage" : "https://openpipelines.bio",
+      "documentation" : "https://openpipelines.bio/fundamentals",
+      "issue_tracker" : "https://github.com/openpipelines-bio/openpipeline/issues"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { filter_with_counts } from "${meta.resources_dir}/../../../../nextflow/filter/filter_with_counts/main.nf"
+include { filter_with_scrublet } from "${meta.resources_dir}/../../../../nextflow/filter/filter_with_scrublet/main.nf"
+include { do_filter } from "${meta.resources_dir}/../../../../nextflow/filter/do_filter/main.nf"
+include { delimit_fraction } from "${meta.resources_dir}/../../../../nextflow/filter/delimit_fraction/main.nf"
+include { qc } from "${meta.resources_dir}/../../../../nextflow/workflows/qc/qc/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+  input_ch
+
+  main:
+  output_ch = input_ch
+    // Set aside the output for this workflow to avoid conflicts
+    | map {id, state -> 
+      def new_state = state + ["workflow_output": state.output]
+      [id, new_state]
+    }
+    // Check for correctness of mitochondrial and ribosomal gene detection arguments
+    | map { id, state ->
+      def new_state = [:]
+      if (state.obs_name_mitochondrial_fraction && !state.var_name_mitochondrial_genes) {
+        throw new RuntimeException("Using --obs_name_mitochondrial_fraction requires --var_name_mitochondrial_genes.")
+      }
+      if (!state.obs_name_mitochondrial_fraction && state.var_name_mitochondrial_genes) {
+        new_state.obs_name_mitochondrial_fraction = "fraction_${state.var_name_mitochondrial_genes}"
+      }
+      if ((state.min_fraction_mito != null || state.max_fraction_mito != null) && !state.var_name_mitochondrial_genes) {
+        throw new RuntimeException("Enabling --min_fraction_mito or --max_fraction_mito requires --var_name_mitochondrial_genes.")
+      }
+      if (state.var_gene_names && !state.var_name_mitochondrial_genes && !state.var_name_ribosomal_genes) {
+        System.err.println("Warning: --var_gene_names is only required for mitochondrial gene detection and does nothing while \
+                           not also setting --var_name_mitochondrial_genes or --var_name_ribosomal_genes.")
+      }
+      if (state.mitochondrial_gene_regex && !state.var_name_mitochondrial_genes && !state.var_name_ribosomal_genes) {
+        System.err.println("Warning: --mitochondrial_gene_regex is only required for mitochondrial gene detection and does \
+                           nothing while not also setting --var_name_mitochondrial_genes or --var_name_ribosomal_genes.")
+      }
+      if (state.obs_name_ribosomal_fraction && !state.var_name_ribosomal_genes) {
+        throw new RuntimeException("Using --obs_name_ribosomal_fraction requires --var_name_ribosomal_genes.")
+      }
+      if (!state.obs_name_ribosomal_fraction && state.var_name_ribosomal_genes) {
+        new_state.obs_name_ribosomal_fraction = "fraction_${state.var_name_ribosomal_genes}"
+      }
+      if ((state.min_fraction_ribo != null || state.max_fraction_ribo != null) && !state.var_name_ribosomal_genes) {
+        throw new RuntimeException("Enabling --min_fraction_ribo or --max_fraction_ribo requires --var_name_ribosomal_genes.")
+      }
+      [id, state + new_state]
+    }
+    | qc.run(
+      fromState: { id, state ->
+        // The rna singlesample processing allows detecting mitochondrial genes and filtering based
+        // on the fraction of mitochondrial genes per cell
+        // This behaviour is optional based on the presence of var_name_mitochondrial_genes
+        // The behavior of other components must be tuned to this argument as well
+        def args = [
+          "id": id,
+          "input": state.input,
+          // disable other qc metric calculations
+          // only mitochondrial gene detection is required at this point
+          "top_n_vars": [],
+          "output_obs_num_nonzero_vars": null,
+          "output_obs_total_counts_vars": null,
+          "output_var_num_nonzero_obs": null,
+          "output_var_total_counts_obs": null,
+          "output_var_obs_mean": null,
+          "output_var_pct_dropout": null,
+          "output": state.output,
+          "modality": "rna",
+          "layer": state.layer,
+        ]
+        
+        if (state.var_name_mitochondrial_genes) {
+          // Check if user has defined var columns to calculate metrics
+          def new_var_qc_metrics = state.var_qc_metrics != null ? state.var_qc_metrics : []
+          assert new_var_qc_metrics instanceof List
+          // Add the mitochondrial genes var column to the columns to calculate statistics for if set.
+          new_var_qc_metrics = ((new_var_qc_metrics as Set) + [state.var_name_mitochondrial_genes]) as List
+          
+          args += [
+            "var_qc_metrics": new_var_qc_metrics,
+            "obs_name_mitochondrial_fraction": state.obs_name_mitochondrial_fraction,
+            "var_gene_names": state.var_gene_names,
+            "var_name_mitochondrial_genes": state.var_name_mitochondrial_genes,
+            "mitochondrial_gene_regex": state.mitochondrial_gene_regex
+          ]
+        }
+
+        if (state.var_name_ribosomal_genes) {
+          // Check if user has defined var columns to calculate metrics
+          def new_var_qc_metrics = state.var_qc_metrics != null ? state.var_qc_metrics : []
+          assert new_var_qc_metrics instanceof List
+          // Add the ribosomal genes var column to the columns to calculate statistics for if set.
+          new_var_qc_metrics = ((new_var_qc_metrics as Set) + [state.var_name_ribosomal_genes]) as List
+          
+          args += [
+            "var_qc_metrics": new_var_qc_metrics,
+            "obs_name_ribosomal_fraction": state.obs_name_ribosomal_fraction,
+            "var_gene_names": state.var_gene_names,
+            "var_name_ribosomal_genes": state.var_name_ribosomal_genes,
+            "ribosomal_gene_regex": state.ribosomal_gene_regex
+          ]
+        }
+
+        return args
+      },
+      toState: ["input": "output"]
+    )
+    | delimit_fraction.run(
+      runIf: {id, state -> state.var_name_mitochondrial_genes},
+      fromState: {id, state -> 
+      [
+        "input": state.input,
+        "obs_name_filter": "filter_mitochondrial",
+        "min_fraction": state.min_fraction_mito,
+        "max_fraction": state.max_fraction_mito,
+        "obs_fraction_column": state.obs_name_mitochondrial_fraction,
+      ]
+      },
+      toState: ["input": "output"]
+    )
+    | delimit_fraction.run(
+      runIf: {id, state -> state.var_name_ribosomal_genes},
+      fromState: {id, state -> 
+      [
+        "input": state.input,
+        "obs_name_filter": "filter_ribosomal",
+        "min_fraction": state.min_fraction_ribo,
+        "max_fraction": state.max_fraction_ribo,
+        "obs_fraction_column": state.obs_name_ribosomal_fraction,
+      ]
+      },
+      toState: ["input": "output"]
+    )
+    // cell filtering
+    | filter_with_counts.run(
+      key: "rna_filter_with_counts",
+      fromState: { id, state ->
+        [
+          "input": state.input,
+          "layer": state.layer,
+          "obs_name_filter": "filter_with_counts",
+          "var_name_filter": "filter_with_counts",
+          "min_counts": state.min_counts,
+          "max_counts": state.max_counts,
+          "min_genes_per_cell": state.min_genes_per_cell,
+          "max_genes_per_cell": state.max_genes_per_cell,
+          "min_cells_per_gene": state.min_cells_per_gene,
+        ]
+      },
+      toState: ["input": "output"]
+    )
+    | do_filter.run(
+      key: "rna_do_filter",
+      fromState: {id, state ->
+        // do_filter does not need a layer argument because it filters all layers
+        // from a modality.
+        def stateMapping = [
+          input: state.input,
+          var_filter: ["filter_with_counts"],
+          // If scrublet is skipped, the output should be set to the workflow output
+          output: state.workflow_output
+        ]
+        def obs_filter = ["filter_with_counts"]
+        if (state.var_name_mitochondrial_genes) {
+          obs_filter += ["filter_mitochondrial"]
+        }
+        if (state.var_name_ribosomal_genes) {
+          obs_filter += ["filter_ribosomal"]
+        }
+        stateMapping += ["obs_filter": obs_filter]
+        return stateMapping
+      },
+      toState: ["input": "output"]
+    )
+    // doublet calling
+    | filter_with_scrublet.run(
+      runIf: { id, state ->
+        !state.skip_scrublet_doublet_detection
+      },
+      fromState: [
+        "input": "input",
+        "layer": "layer",
+        "output": "workflow_output"
+      ],
+      args: [output_compression: "gzip"],
+      toState: ["input": "output"]
+    )
+    | setState(["output": "input"])
+
+  emit:
+  output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/nextflow.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/nextflow.config
new file mode 100644
index 0000000..9856e3d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/rna/rna_singlesample'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v4.0.4'
+  description = 'Processing unimodal single-sample RNA transcriptomics data.'
+  author = 'Dries De Maeyer, Robrecht Cannoodt, Dries Schaumont'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/nextflow_labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/nextflow_labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/nextflow_schema.json
new file mode 100644
index 0000000..876b7f5
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/nextflow_schema.json
@@ -0,0 +1,177 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "rna_singlesample",
+  "description": "Processing unimodal single-sample RNA transcriptomics data.",
+  "type": "object",
+  "$defs": {
+    "input": {
+      "title": "Input",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "ID of the sample.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"foo\"`. "
+        },
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Path to the sample.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"dataset.h5mu\"`. "
+        },
+        "layer": {
+          "type": "string",
+          "description": "Input layer to start from",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "output": {
+      "title": "Output",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "Destination path to the output.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "filtering options": {
+      "title": "Filtering options",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "min_counts": {
+          "type": "integer",
+          "description": "Minimum number of counts captured per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "max_counts": {
+          "type": "integer",
+          "description": "Maximum number of counts captured per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `5000000`. "
+        },
+        "min_genes_per_cell": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "max_genes_per_cell": {
+          "type": "integer",
+          "description": "Maximum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `1500000`. "
+        },
+        "min_cells_per_gene": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per gene.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3`. "
+        },
+        "min_fraction_mito": {
+          "type": "number",
+          "description": "Minimum fraction of UMIs that are mitochondrial",
+          "help_text": "Type: `double`, multiple: `False`, example: `0.0`. "
+        },
+        "max_fraction_mito": {
+          "type": "number",
+          "description": "Maximum fraction of UMIs that are mitochondrial",
+          "help_text": "Type: `double`, multiple: `False`, example: `0.2`. "
+        },
+        "min_fraction_ribo": {
+          "type": "number",
+          "description": "Minimum fraction of UMIs that are ribosomal",
+          "help_text": "Type: `double`, multiple: `False`, example: `0.0`. "
+        },
+        "max_fraction_ribo": {
+          "type": "number",
+          "description": "Maximum fraction of UMIs that are ribosomal",
+          "help_text": "Type: `double`, multiple: `False`, example: `0.2`. "
+        },
+        "skip_scrublet_doublet_detection": {
+          "type": "boolean",
+          "description": "Skip the scrublet doublet detection step.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "mitochondrial & ribosomal gene detection": {
+      "title": "Mitochondrial & Ribosomal Gene Detection",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "var_gene_names": {
+          "type": "string",
+          "description": ".var column name to be used to detect mitochondrial/ribosomal genes instead of .var_names (default if not set).\nGene names matching with the regex value from --mitochondrial_gene_regex or --ribosomal_gene_regex will be \nidentified as mitochondrial or ribosomal genes, respectively",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"gene_symbol\"`. "
+        },
+        "var_name_mitochondrial_genes": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_name_mitochondrial_fraction": {
+          "type": "string",
+          "description": "When specified, write the fraction of counts originating from mitochondrial genes \n(based on --mitochondrial_gene_regex) to an .obs column with the specified name.\nRequires --var_name_mitochondrial_genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "mitochondrial_gene_regex": {
+          "type": "string",
+          "description": "Regex string that identifies mitochondrial genes from --var_gene_names.\nBy default will detect human and mouse mitochondrial genes from a gene symbol.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"^[mM][tT]-\"`. ",
+          "default": "^[mM][tT]-"
+        },
+        "var_name_ribosomal_genes": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_name_ribosomal_fraction": {
+          "type": "string",
+          "description": "When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified name.\nRequires --var_name_ribosomal_genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "ribosomal_gene_regex": {
+          "type": "string",
+          "description": "Regex string that identifies ribosomal genes from --var_gene_names.\nBy default will detect human and mouse ribosomal genes from a gene symbol.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"^[Mm]?[Rr][Pp][LlSs]\"`. ",
+          "default": "^[Mm]?[Rr][Pp][LlSs]"
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/input"
+    },
+    {
+      "$ref": "#/$defs/output"
+    },
+    {
+      "$ref": "#/$defs/filtering options"
+    },
+    {
+      "$ref": "#/$defs/mitochondrial & ribosomal gene detection"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/utils/errorstrat_ignore.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/utils/errorstrat_ignore.config
new file mode 100644
index 0000000..6b4b029
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/utils/errorstrat_ignore.config
@@ -0,0 +1 @@
+process.errorStrategy = 'ignore' 
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/utils/integration_tests.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/utils/integration_tests.config
new file mode 100644
index 0000000..59d5b09
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/utils/integration_tests.config
@@ -0,0 +1,36 @@
+profiles {
+
+  // detect tempdir
+  tempDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?:
+      System.getenv('VIASH_TEMP') ?: 
+      System.getenv('TEMPDIR') ?: 
+      System.getenv('TMPDIR') ?: 
+      '/tmp'
+  ).toAbsolutePath()
+
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/utils/labels.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/utils/labels.config
new file mode 100644
index 0000000..94570ec
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/utils/labels.config
@@ -0,0 +1,48 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+
+  // The memory a task is assinged increases with each attempt
+  // uncomment the line below and adjust the value to set a global upper limit on the memory.
+  // resourceLimits = [ memory: 240.Gb ] 
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: lowmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 4.GB * task.attempt } }
+  withLabel: midmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 25.GB * task.attempt } }
+  withLabel: highmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 50.GB * task.attempt } }
+  withLabel: veryhighmem { memory = { task?.resourceLimits?.memory && task?.maxRetries && task.attempt >= task.maxRetries ? task.resourceLimits.memory : 75.GB * task.attempt } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/utils/labels_ci.config b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/utils/labels_ci.config
new file mode 100644
index 0000000..aee866d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/rna/rna_singlesample/utils/labels_ci.config
@@ -0,0 +1,33 @@
+process {
+  withLabel: lowmem { memory = 13.Gb }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midmem { memory = 13.Gb }
+  withLabel: midcpu { cpus = 4 }
+  withLabel: highmem { memory = 13.Gb }
+  withLabel: highcpu { cpus = 4 }
+  withLabel: veryhighmem { memory = 13.Gb }
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+}
+
+env.NUMBA_CACHE_DIR = '/tmp'
+
+trace {
+    enabled = true
+    overwrite = true
+}
+dag {
+  overwrite = true
+}
+
+process.maxForks = 1
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/.config.vsh.yaml
new file mode 100644
index 0000000..59166f4
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/.config.vsh.yaml
@@ -0,0 +1,563 @@
+name: "h5mu_to_qc_json"
+namespace: "ingestion_qc"
+version: "v0.2.2"
+authors:
+- name: "Jakub Majercik"
+  roles:
+  - "author"
+  info:
+    role: "Contributor"
+    links:
+      email: "jakub@data-intuitive.com"
+      github: "jakubmajercik"
+      linkedin: "jakubmajercik"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Bioinformatics Engineer"
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "The input h5mu file(s)"
+    info: null
+    example:
+    - "path/to/file.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "The modality to use"
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ingestion_method"
+    description: "Method that was used to ingest the data - this will define the structure\
+      \ of the report that is generated."
+    info: null
+    required: true
+    choices:
+    - "cellranger_multi"
+    - "xenium"
+    - "cosmx"
+    - "visium"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_sample_id"
+    description: "The key in the h5mu file that contains the sample ID. If not provided,\
+      \ each H5MU file will be considered as a separate sample."
+    info: null
+    default:
+    - "sample_id"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_total_counts"
+    description: "The key in the h5mu .obs field that contains the total counts."
+    info: null
+    default:
+    - "total_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_num_nonzero_vars"
+    description: "The key in the h5mu .obs field that contains the number of nonzero\
+      \ vars."
+    info: null
+    default:
+    - "num_nonzero_vars"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_fraction_mitochondrial"
+    description: "The key in the h5mu .obs field that contains the fraction mitochondrial\
+      \ genes."
+    info: null
+    default:
+    - "fraction_mitochondrial"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_fraction_ribosomal"
+    description: "The key in the h5mu .obs field that contains the fraction ribosomal\
+      \ genes."
+    info: null
+    default:
+    - "fraction_ribosomal"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "The output JSON file"
+    info: null
+    example:
+    - "path/to/file.json"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_reporting_json"
+    description: "The output JSON file that defines the QC report"
+    info: null
+    example:
+    - "path/to/file.json"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "Filtering & grouping options"
+  arguments:
+  - type: "integer"
+    name: "--min_total_counts"
+    description: "Minimum total counts for a cell to be included in the output"
+    info: null
+    default:
+    - 10
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_num_nonzero_vars"
+    description: "Minimum number of nonzero vars for a cell to be included in the\
+      \ output"
+    info: null
+    default:
+    - 10
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_metadata"
+    description: "The metadata keys in the h5mu .obs to include in the output JSON."
+    info: null
+    example:
+    - "donor_id;cell_type;batch;condition"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Options for CellRanger reports"
+  arguments:
+  - type: "string"
+    name: "--obs_cellbender"
+    description: "The cellbender keys in the h5mu .obs to include in the output JSON"
+    info: null
+    default:
+    - "cellbender_background_fraction"
+    - "cellbender_cell_probability"
+    - "cellbender_cell_size"
+    - "cellbender_droplet_efficiency"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--uns_cellranger_metrics"
+    description: "The key in the h5mu file .uns that contains the cellranger metrics"
+    info: null
+    default:
+    - "metrics_cellranger"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Options for Xenium reports"
+  arguments:
+  - type: "string"
+    name: "--obs_nucleus_area"
+    description: "The key in the h5mu .obs field that contains the nucleus area."
+    info: null
+    default:
+    - "nucleus_area"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_cell_area"
+    description: "The key in the h5mu .obs field that contains the cell area."
+    info: null
+    default:
+    - "cell_area"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_x_coord"
+    description: "The key in the h5mu .obs field that contains the x coordinate."
+    info: null
+    default:
+    - "x_coord"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_y_coord"
+    description: "The key in the h5mu .obs field that contains the y coordinate."
+    info: null
+    default:
+    - "y_coord"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_control_probe_counts"
+    description: "The key in the h5mu .obs field that contains the number of control\
+      \ probes."
+    info: null
+    default:
+    - "control_probe_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_control_codeword_counts"
+    description: "The key in the h5mu .obs field that contains the number of control\
+      \ codewords."
+    info: null
+    default:
+    - "control_codeword_counts"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Options for CosMx reports"
+  arguments:
+  - type: "string"
+    name: "--obs_area"
+    description: "The key in the h5mu .obs field that contains the cell area."
+    info: null
+    default:
+    - "Area"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_aspect_ratio"
+    description: "The key in the h5mu .obs field that contains the aspect ratio."
+    info: null
+    default:
+    - "AspectRatio"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_dapi_stain"
+    description: "The keys in the h5mu .obs field that contains the DAPI stain intensity."
+    info: null
+    default:
+    - "Mean.DAPI"
+    - "Max.DAPI"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_membrane_stain"
+    description: "The keys in the h5mu .obs field that contains the membrane stain\
+      \ intensity."
+    info: null
+    default:
+    - "Mean.MembraneStain"
+    - "Max.MembraneStain"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_panck_stain"
+    description: "The keys in the h5mu .obs field that contains the PanCK stain intensity."
+    info: null
+    default:
+    - "Mean.PanCK"
+    - "Max.PanCK"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_cd45_stain"
+    description: "The keys in the h5mu .obs field that contains the CD45 stain intensity."
+    info: null
+    default:
+    - "Mean.CD45"
+    - "Max.CD45"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_cd3_stain"
+    description: "The keys in the h5mu .obs field that contains the CD3 stain intensity."
+    info: null
+    default:
+    - "Mean.CD3"
+    - "Max.CD3"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+resources:
+- type: "python_script"
+  path: "script.py"
+  is_executable: true
+- type: "file"
+  path: "setup_logger.py"
+- type: "file"
+  path: "report_structure"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Takes H5MU files that have been ingested by CellRanger, Xenium or CosMx\
+  \ and processed by the QC workflow, and generates:\n- A JSON file that contains\
+  \ the combined data for the QC report\n- A JSON file that defines the layout and\
+  \ structure of the QC report\n"
+test_resources:
+- type: "python_script"
+  path: "test.py"
+  is_executable: true
+- type: "file"
+  path: "resources_test"
+info: null
+status: "enabled"
+scope:
+  image: "private"
+  target: "private"
+requirements:
+  commands:
+  - "ps"
+repositories:
+- type: "vsh"
+  name: "openpipeline"
+  repo: "openpipeline"
+  tag: "v4.0.0"
+- type: "vsh"
+  name: "craftbox"
+  repo: "craftbox"
+  tag: "v0.2.0"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline_qc"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "midmem"
+    - "middisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "python:3.12-slim"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v0.2.2"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "procps"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "anndata~=0.11.1"
+    - "mudata~=0.3.1"
+    script:
+    - "exec(\"try:\\n  import awkward\\nexcept ModuleNotFoundError:\\n  exit(0)\\\
+      nelse:  exit(1)\")"
+    upgrade: true
+  test_setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "python"
+    user: false
+    packages:
+    - "viashpy==0.8.0"
+    github:
+    - "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+    upgrade: true
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/ingestion_qc/h5mu_to_qc_json/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/_private/nextflow/ingestion_qc/h5mu_to_qc_json"
+  executable: "target/_private/nextflow/ingestion_qc/h5mu_to_qc_json/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "9b81d310d031b120755e39001cc40f8ec51bc7c2"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline_qc"
+package_config:
+  name: "openpipeline_qc"
+  version: "v0.2.2"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-bio/openpipeline_incubator/resources_test"
+      dest: "resources_test"
+  repositories:
+  - type: "vsh"
+    name: "openpipeline"
+    repo: "openpipeline"
+    tag: "v4.0.0"
+  - type: "vsh"
+    name: "craftbox"
+    repo: "craftbox"
+    tag: "v0.2.0"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag\
+    \ := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v0.2.2'"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline_qc"
+    docker_registry: "ghcr.io"
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/main.nf b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/main.nf
new file mode 100644
index 0000000..3630316
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/main.nf
@@ -0,0 +1,4768 @@
+// h5mu_to_qc_json v0.2.2
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Jakub Majercik (author)
+//  * Dorien Roosen (author)
+//  * Robrecht Cannoodt (author)
+//  * Weiwei Schultz (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "h5mu_to_qc_json",
+  "namespace" : "ingestion_qc",
+  "version" : "v0.2.2",
+  "authors" : [
+    {
+      "name" : "Jakub Majercik",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "links" : {
+          "email" : "jakub@data-intuitive.com",
+          "github" : "jakubmajercik",
+          "linkedin" : "jakubmajercik"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Bioinformatics Engineer"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "The input h5mu file(s)",
+          "example" : [
+            "path/to/file.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "The modality to use",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--ingestion_method",
+          "description" : "Method that was used to ingest the data - this will define the structure of the report that is generated.",
+          "required" : true,
+          "choices" : [
+            "cellranger_multi",
+            "xenium",
+            "cosmx",
+            "visium"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_sample_id",
+          "description" : "The key in the h5mu file that contains the sample ID. If not provided, each H5MU file will be considered as a separate sample.",
+          "default" : [
+            "sample_id"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_total_counts",
+          "description" : "The key in the h5mu .obs field that contains the total counts.",
+          "default" : [
+            "total_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_num_nonzero_vars",
+          "description" : "The key in the h5mu .obs field that contains the number of nonzero vars.",
+          "default" : [
+            "num_nonzero_vars"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_fraction_mitochondrial",
+          "description" : "The key in the h5mu .obs field that contains the fraction mitochondrial genes.",
+          "default" : [
+            "fraction_mitochondrial"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_fraction_ribosomal",
+          "description" : "The key in the h5mu .obs field that contains the fraction ribosomal genes.",
+          "default" : [
+            "fraction_ribosomal"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "The output JSON file",
+          "example" : [
+            "path/to/file.json"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_reporting_json",
+          "description" : "The output JSON file that defines the QC report",
+          "example" : [
+            "path/to/file.json"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Filtering & grouping options",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--min_total_counts",
+          "description" : "Minimum total counts for a cell to be included in the output",
+          "default" : [
+            10
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_num_nonzero_vars",
+          "description" : "Minimum number of nonzero vars for a cell to be included in the output",
+          "default" : [
+            10
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_metadata",
+          "description" : "The metadata keys in the h5mu .obs to include in the output JSON.",
+          "example" : [
+            "donor_id;cell_type;batch;condition"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Options for CellRanger reports",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obs_cellbender",
+          "description" : "The cellbender keys in the h5mu .obs to include in the output JSON",
+          "default" : [
+            "cellbender_background_fraction",
+            "cellbender_cell_probability",
+            "cellbender_cell_size",
+            "cellbender_droplet_efficiency"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--uns_cellranger_metrics",
+          "description" : "The key in the h5mu file .uns that contains the cellranger metrics",
+          "default" : [
+            "metrics_cellranger"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Options for Xenium reports",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obs_nucleus_area",
+          "description" : "The key in the h5mu .obs field that contains the nucleus area.",
+          "default" : [
+            "nucleus_area"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_cell_area",
+          "description" : "The key in the h5mu .obs field that contains the cell area.",
+          "default" : [
+            "cell_area"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_x_coord",
+          "description" : "The key in the h5mu .obs field that contains the x coordinate.",
+          "default" : [
+            "x_coord"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_y_coord",
+          "description" : "The key in the h5mu .obs field that contains the y coordinate.",
+          "default" : [
+            "y_coord"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_control_probe_counts",
+          "description" : "The key in the h5mu .obs field that contains the number of control probes.",
+          "default" : [
+            "control_probe_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_control_codeword_counts",
+          "description" : "The key in the h5mu .obs field that contains the number of control codewords.",
+          "default" : [
+            "control_codeword_counts"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Options for CosMx reports",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obs_area",
+          "description" : "The key in the h5mu .obs field that contains the cell area.",
+          "default" : [
+            "Area"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_aspect_ratio",
+          "description" : "The key in the h5mu .obs field that contains the aspect ratio.",
+          "default" : [
+            "AspectRatio"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_dapi_stain",
+          "description" : "The keys in the h5mu .obs field that contains the DAPI stain intensity.",
+          "default" : [
+            "Mean.DAPI",
+            "Max.DAPI"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_membrane_stain",
+          "description" : "The keys in the h5mu .obs field that contains the membrane stain intensity.",
+          "default" : [
+            "Mean.MembraneStain",
+            "Max.MembraneStain"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_panck_stain",
+          "description" : "The keys in the h5mu .obs field that contains the PanCK stain intensity.",
+          "default" : [
+            "Mean.PanCK",
+            "Max.PanCK"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_cd45_stain",
+          "description" : "The keys in the h5mu .obs field that contains the CD45 stain intensity.",
+          "default" : [
+            "Mean.CD45",
+            "Max.CD45"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_cd3_stain",
+          "description" : "The keys in the h5mu .obs field that contains the CD3 stain intensity.",
+          "default" : [
+            "Mean.CD3",
+            "Max.CD3"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "python_script",
+      "path" : "script.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/utils/setup_logger.py"
+    },
+    {
+      "type" : "file",
+      "path" : "report_structure"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/configs/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Takes H5MU files that have been ingested by CellRanger, Xenium or CosMx and processed by the QC workflow, and generates:\n- A JSON file that contains the combined data for the QC report\n- A JSON file that defines the layout and structure of the QC report\n",
+  "test_resources" : [
+    {
+      "type" : "python_script",
+      "path" : "test.py",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "private",
+    "target" : "private"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "repositories" : [
+    {
+      "type" : "vsh",
+      "name" : "openpipeline",
+      "repo" : "openpipeline",
+      "tag" : "v4.0.0"
+    },
+    {
+      "type" : "vsh",
+      "name" : "craftbox",
+      "repo" : "craftbox",
+      "tag" : "v0.2.0"
+    }
+  ],
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline_qc",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "midmem",
+          "middisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "python:3.12-slim",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v0.2.2",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "procps"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "anndata~=0.11.1",
+            "mudata~=0.3.1"
+          ],
+          "script" : [
+            "exec(\\"try:\\\\n  import awkward\\\\nexcept ModuleNotFoundError:\\\\n  exit(0)\\\\nelse:  exit(1)\\")"
+          ],
+          "upgrade" : true
+        }
+      ],
+      "test_setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "python",
+          "user" : false,
+          "packages" : [
+            "viashpy==0.8.0"
+          ],
+          "github" : [
+            "openpipelines-bio/core#subdirectory=packages/python/openpipeline_testutils"
+          ],
+          "upgrade" : true
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/ingestion_qc/h5mu_to_qc_json/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/_private/nextflow/ingestion_qc/h5mu_to_qc_json",
+    "viash_version" : "0.9.4",
+    "git_commit" : "9b81d310d031b120755e39001cc40f8ec51bc7c2",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline_qc"
+  },
+  "package_config" : {
+    "name" : "openpipeline_qc",
+    "version" : "v0.2.2",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-bio/openpipeline_incubator/resources_test",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "repositories" : [
+      {
+        "type" : "vsh",
+        "name" : "openpipeline",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.0"
+      },
+      {
+        "type" : "vsh",
+        "name" : "craftbox",
+        "repo" : "craftbox",
+        "tag" : "v0.2.0"
+      }
+    ],
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v0.2.2'"
+    ],
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline_qc",
+      "docker_registry" : "ghcr.io"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.py"
+cat > "$tempscript" << VIASHMAIN
+import json
+import pandas as pd
+from pathlib import Path
+import anndata as ad
+import h5py
+import sys
+import os
+import shutil
+
+## VIASH START
+# The following code has been auto-generated by Viash.
+par = {
+  'input': $( if [ ! -z ${VIASH_PAR_INPUT+x} ]; then echo "r'${VIASH_PAR_INPUT//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'modality': $( if [ ! -z ${VIASH_PAR_MODALITY+x} ]; then echo "r'${VIASH_PAR_MODALITY//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'ingestion_method': $( if [ ! -z ${VIASH_PAR_INGESTION_METHOD+x} ]; then echo "r'${VIASH_PAR_INGESTION_METHOD//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_sample_id': $( if [ ! -z ${VIASH_PAR_OBS_SAMPLE_ID+x} ]; then echo "r'${VIASH_PAR_OBS_SAMPLE_ID//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_total_counts': $( if [ ! -z ${VIASH_PAR_OBS_TOTAL_COUNTS+x} ]; then echo "r'${VIASH_PAR_OBS_TOTAL_COUNTS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_num_nonzero_vars': $( if [ ! -z ${VIASH_PAR_OBS_NUM_NONZERO_VARS+x} ]; then echo "r'${VIASH_PAR_OBS_NUM_NONZERO_VARS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_fraction_mitochondrial': $( if [ ! -z ${VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL+x} ]; then echo "r'${VIASH_PAR_OBS_FRACTION_MITOCHONDRIAL//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_fraction_ribosomal': $( if [ ! -z ${VIASH_PAR_OBS_FRACTION_RIBOSOMAL+x} ]; then echo "r'${VIASH_PAR_OBS_FRACTION_RIBOSOMAL//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output': $( if [ ! -z ${VIASH_PAR_OUTPUT+x} ]; then echo "r'${VIASH_PAR_OUTPUT//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'output_reporting_json': $( if [ ! -z ${VIASH_PAR_OUTPUT_REPORTING_JSON+x} ]; then echo "r'${VIASH_PAR_OUTPUT_REPORTING_JSON//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'min_total_counts': $( if [ ! -z ${VIASH_PAR_MIN_TOTAL_COUNTS+x} ]; then echo "int(r'${VIASH_PAR_MIN_TOTAL_COUNTS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'min_num_nonzero_vars': $( if [ ! -z ${VIASH_PAR_MIN_NUM_NONZERO_VARS+x} ]; then echo "int(r'${VIASH_PAR_MIN_NUM_NONZERO_VARS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'obs_metadata': $( if [ ! -z ${VIASH_PAR_OBS_METADATA+x} ]; then echo "r'${VIASH_PAR_OBS_METADATA//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'obs_cellbender': $( if [ ! -z ${VIASH_PAR_OBS_CELLBENDER+x} ]; then echo "r'${VIASH_PAR_OBS_CELLBENDER//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'uns_cellranger_metrics': $( if [ ! -z ${VIASH_PAR_UNS_CELLRANGER_METRICS+x} ]; then echo "r'${VIASH_PAR_UNS_CELLRANGER_METRICS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_nucleus_area': $( if [ ! -z ${VIASH_PAR_OBS_NUCLEUS_AREA+x} ]; then echo "r'${VIASH_PAR_OBS_NUCLEUS_AREA//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_cell_area': $( if [ ! -z ${VIASH_PAR_OBS_CELL_AREA+x} ]; then echo "r'${VIASH_PAR_OBS_CELL_AREA//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_x_coord': $( if [ ! -z ${VIASH_PAR_OBS_X_COORD+x} ]; then echo "r'${VIASH_PAR_OBS_X_COORD//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_y_coord': $( if [ ! -z ${VIASH_PAR_OBS_Y_COORD+x} ]; then echo "r'${VIASH_PAR_OBS_Y_COORD//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_control_probe_counts': $( if [ ! -z ${VIASH_PAR_OBS_CONTROL_PROBE_COUNTS+x} ]; then echo "r'${VIASH_PAR_OBS_CONTROL_PROBE_COUNTS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_control_codeword_counts': $( if [ ! -z ${VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS+x} ]; then echo "r'${VIASH_PAR_OBS_CONTROL_CODEWORD_COUNTS//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_area': $( if [ ! -z ${VIASH_PAR_OBS_AREA+x} ]; then echo "r'${VIASH_PAR_OBS_AREA//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_aspect_ratio': $( if [ ! -z ${VIASH_PAR_OBS_ASPECT_RATIO+x} ]; then echo "r'${VIASH_PAR_OBS_ASPECT_RATIO//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'obs_dapi_stain': $( if [ ! -z ${VIASH_PAR_OBS_DAPI_STAIN+x} ]; then echo "r'${VIASH_PAR_OBS_DAPI_STAIN//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'obs_membrane_stain': $( if [ ! -z ${VIASH_PAR_OBS_MEMBRANE_STAIN+x} ]; then echo "r'${VIASH_PAR_OBS_MEMBRANE_STAIN//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'obs_panck_stain': $( if [ ! -z ${VIASH_PAR_OBS_PANCK_STAIN+x} ]; then echo "r'${VIASH_PAR_OBS_PANCK_STAIN//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'obs_cd45_stain': $( if [ ! -z ${VIASH_PAR_OBS_CD45_STAIN+x} ]; then echo "r'${VIASH_PAR_OBS_CD45_STAIN//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi ),
+  'obs_cd3_stain': $( if [ ! -z ${VIASH_PAR_OBS_CD3_STAIN+x} ]; then echo "r'${VIASH_PAR_OBS_CD3_STAIN//\\'/\\'\\"\\'\\"r\\'}'.split(';')"; else echo None; fi )
+}
+meta = {
+  'name': $( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "r'${VIASH_META_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'functionality_name': $( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "r'${VIASH_META_FUNCTIONALITY_NAME//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'resources_dir': $( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "r'${VIASH_META_RESOURCES_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'executable': $( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "r'${VIASH_META_EXECUTABLE//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'config': $( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "r'${VIASH_META_CONFIG//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'temp_dir': $( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "r'${VIASH_META_TEMP_DIR//\\'/\\'\\"\\'\\"r\\'}'"; else echo None; fi ),
+  'cpus': $( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "int(r'${VIASH_META_CPUS//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_b': $( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "int(r'${VIASH_META_MEMORY_B//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kb': $( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mb': $( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gb': $( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tb': $( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pb': $( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_kib': $( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_KIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_mib': $( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_MIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_gib': $( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_GIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_tib': $( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_TIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi ),
+  'memory_pib': $( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "int(r'${VIASH_META_MEMORY_PIB//\\'/\\'\\"\\'\\"r\\'}')"; else echo None; fi )
+}
+dep = {
+  
+}
+
+## VIASH END
+
+sys.path.append(meta["resources_dir"])
+from setup_logger import setup_logger
+
+logger = setup_logger()
+
+par["obs_cellbender"] = {} if not par["obs_cellbender"] else par["obs_cellbender"]
+par["obs_dapi_stain"] = [] if not par["obs_dapi_stain"] else par["obs_dapi_stain"]
+par["obs_membrane_stain"] = [] if not par["obs_membrane_stain"] else par["obs_membrane_stain"]
+par["obs_panck_stain"] = [] if not par["obs_panck_stain"] else par["obs_panck_stain"]
+par["obs_cd45_stain"] = [] if not par["obs_cd45_stain"] else par["obs_cd45_stain"]
+par["obs_cd3_stain"] = [] if not par["obs_cd3_stain"] else par["obs_cd3_stain"]
+
+
+def transform_df(df):
+    """Transform a DataFrame into the annotation object format."""
+    columns = []
+    for name in df.columns:
+        data = df[name]
+
+        # Determine dtype
+        if pd.api.types.is_integer_dtype(data):
+            dtype = "integer"
+        elif pd.api.types.is_float_dtype(data):
+            dtype = "numeric"
+        elif pd.api.types.is_categorical_dtype(data):
+            dtype = "categorical"
+        else:
+            raise ValueError(f"Unknown/unsupported data type for column {name}")
+
+        column_info = {"name": name, "dtype": dtype}
+
+        if dtype == "categorical":
+            column_info["data"] = data.cat.codes.tolist()
+            column_info["categories"] = data.cat.categories.tolist()
+        else:
+            column_info["data"] = [None if pd.isna(x) else x for x in data]
+
+        columns.append(column_info)
+
+    return {"num_rows": len(df), "num_cols": len(df.columns), "min_total_counts": par["min_total_counts"], "min_num_nonzero_vars": par["min_num_nonzero_vars"], "columns": columns}
+
+
+def check_optional_obs_keys(obs, keys, message):
+    missing_keys = [key for key in keys if key not in obs.columns]
+    if missing_keys:
+        logger.info(f"Missing keys in obs: {', '.join(missing_keys)}. {message}")
+
+
+def transform_cellranger_metrics(uns, sample_id):
+    if not par["uns_cellranger_metrics"] in uns:
+        raise ValueError(f"Could not find cellranger metrics in uns: {par['uns_cellranger_metrics']}. Provide correct value for --uns_cellranger_metrics or make sure data was ingested using CellRanger multi.")
+
+    cellranger_metrics = (
+        uns[par["uns_cellranger_metrics"]]
+        .pivot_table(
+            index=[],
+            columns="Metric Name",
+            values="Metric Value",
+            aggfunc="first",
+        )
+        .reset_index(drop=True)
+    )
+
+    cellranger_metrics.columns.name = None
+    # Remove thousands separator and convert to numeric
+    cellranger_metrics = cellranger_metrics.map(
+        lambda x: (
+            pd.to_numeric(x.replace(",", ""), errors="coerce")
+            if isinstance(x, str)
+            else x
+        )
+    )
+    # Replace spaces with underscores in column names
+    cellranger_metrics.columns = cellranger_metrics.columns.str.replace(" ", "_")
+    for col in cellranger_metrics.columns:
+        cellranger_metrics[col] = pd.to_numeric(cellranger_metrics[col], errors="coerce")
+    cellranger_metrics["sample_id"] = [sample_id[0]]
+
+    return cellranger_metrics
+
+
+def format_cellbender_columns(mod_obs):
+    # Check if celbender was run on the dataset
+    if par["obs_cellbender"]:
+        check_optional_obs_keys(mod_obs, par["obs_cellbender"], "Run cellbender first to include these metrics.")
+
+    cellbender_obs_keys = [column for column in par["obs_cellbender"] if column in mod_obs]
+
+    for key in cellbender_obs_keys:
+        if not pd.api.types.is_float_dtype(mod_obs[key]):
+            try:
+                mod_obs[key] = mod_obs[key].astype("float16")
+            except ValueError:
+                raise ValueError(f"Could not convert column {key} to a float dtype. Please make sure all cellbender metrics are numeric.")
+
+    return cellbender_obs_keys, mod_obs
+
+
+def format_required_columns(required_keys, mod_obs):
+
+    for key in required_keys:
+        if not pd.api.types.is_numeric_dtype(mod_obs[key]):
+            raise ValueError(f"Column {key} must be a numeric dtype.")
+
+    if not pd.api.types.is_integer_dtype(mod_obs[par["obs_total_counts"]]):
+        logger.info(f"Converting {par['obs_total_counts']} from {mod_obs[par['obs_total_counts']].dtype} to integer dtype...")
+        mod_obs[par["obs_total_counts"]] = mod_obs[par["obs_total_counts"]].astype(int)
+
+    if not pd.api.types.is_integer_dtype(mod_obs[par["obs_num_nonzero_vars"]]):
+        logger.info(f"Converting {par['obs_num_nonzero_vars']} from {mod_obs[par['obs_num_nonzero_vars']].dtype} to integer dtype...")
+        mod_obs[par["obs_num_nonzero_vars"]] = mod_obs[par["obs_num_nonzero_vars"]].astype(int)
+
+    if not pd.api.types.is_float_dtype(mod_obs[par["obs_fraction_mitochondrial"]]):
+        logger.info(f"Converting {par['obs_fraction_mitochondrial']} from {mod_obs[par['obs_fraction_mitochondrial']].dtype} to float dtype...")
+        mod_obs[par["obs_fraction_mitochondrial"]] = mod_obs[par["obs_fraction_mitochondrial"]].astype("float16")
+
+    if not pd.api.types.is_float_dtype(mod_obs[par["obs_fraction_ribosomal"]]):
+        logger.info(f"Converting {par['obs_fraction_ribosomal']} from {mod_obs[par['obs_fraction_ribosomal']].dtype} to float dtype...")
+        mod_obs[par["obs_fraction_ribosomal"]] = mod_obs[par["obs_fraction_ribosomal"]].astype("float16")
+
+    return mod_obs
+
+
+def format_categorical_columns(mod_obs):
+    # Fetch all categorical columns for grouping if no columns are provided
+    if not par["obs_metadata"]:
+        metadata_obs_keys = mod_obs.select_dtypes(include=["object", "category"]).columns.tolist()
+        if par["obs_sample_id"] in metadata_obs_keys:
+            metadata_obs_keys.remove(par["obs_sample_id"])
+    else:
+        check_optional_obs_keys(mod_obs, par["obs_metadata"], "Make sure requested metadata colmuns are present in obs.")
+        metadata_obs_keys = [key for key in par["obs_metadata"] if key in mod_obs]
+
+    for key in metadata_obs_keys:
+        if not isinstance(key, pd.CategoricalDtype):
+            logger.info(f"{key} is not a categorical dtype. Converting {key} from {mod_obs[key].dtype} to categorical dtype...")
+            mod_obs[key] = mod_obs[key].astype(str).astype("category")
+
+    return metadata_obs_keys, mod_obs
+
+
+def generate_cellranger_stats(mod_obs, uns, sample_id, required_keys):
+
+    # Format required columns
+    mod_obs = format_required_columns(required_keys, mod_obs)
+
+    # Fetch and format  all categorical columns for grouping
+    metadata_obs_keys, mod_obs = format_categorical_columns(mod_obs)
+
+    # Fetch and format cellbender columns
+    cellbender_obs_keys, mod_obs = format_cellbender_columns(mod_obs)
+
+    # Create cell RNA stats dataframe
+    cell_rna_stats = pd.DataFrame(
+        {
+            "sample_id": pd.Categorical(sample_id),
+            **{key: mod_obs[key] for key in required_keys},
+            **{key: mod_obs[key] for key in cellbender_obs_keys},
+            **{key: mod_obs[key] for key in metadata_obs_keys},
+        }
+    )
+
+    cellranger_stats = transform_cellranger_metrics(uns, sample_id)
+
+    return cell_rna_stats, cellranger_stats
+
+
+def format_xenium_columns(mod_obs):
+
+    mod_obs["nucleus_ratio"] = mod_obs[par["obs_nucleus_area"]] / mod_obs[par["obs_cell_area"]]
+
+    xenium_formatted_columns = [par["obs_cell_area"], "nucleus_ratio", "x_coord", "y_coord"]
+    for key in xenium_formatted_columns:
+        mod_obs[key] = mod_obs[key].astype("float16")
+
+    return mod_obs, xenium_formatted_columns
+
+
+def generate_xenium_stats(mod_obs, sample_id, required_keys):
+
+    # Format required columns
+    mod_obs = format_required_columns(required_keys, mod_obs)
+
+    # Format xenium-specific columns
+    mod_obs, xenium_formatted_columns = format_xenium_columns(mod_obs)
+
+    # Fetch and format  all categorical columns for grouping
+    metadata_obs_keys, mod_obs = format_categorical_columns(mod_obs)
+
+    # Create cell RNA stats dataframe
+    cell_rna_stats = pd.DataFrame(
+        {
+            "sample_id": pd.Categorical(sample_id),
+            **{key: mod_obs[key] for key in required_keys},
+            **{key: mod_obs[key] for key in xenium_formatted_columns},
+            **{key: mod_obs[key] for key in metadata_obs_keys}
+        }
+    )
+    
+    return cell_rna_stats
+
+def generate_visium_stats(mod_obs, sample_id, required_keys):
+    
+    # Format required columns
+    mod_obs = format_required_columns(required_keys, mod_obs)
+
+    # Format visium-specific columns
+    visium_formatted_columns = ["x_coord", "y_coord"]
+    for key in visium_formatted_columns:
+        mod_obs[key] = mod_obs[key].astype("float16")
+
+    # Fetch and format  all categorical columns for grouping
+    metadata_obs_keys, mod_obs = format_categorical_columns(mod_obs)
+
+    # Create cell RNA stats dataframe
+    cell_rna_stats = pd.DataFrame(
+        {
+            "sample_id": pd.Categorical(sample_id),
+            **{key: mod_obs[key] for key in required_keys},
+            **{key: mod_obs[key] for key in visium_formatted_columns},
+            **{key: mod_obs[key] for key in metadata_obs_keys}
+        }
+    )
+
+    return cell_rna_stats
+
+
+def format_cosmx_columns(mod_obs):
+
+    cosmx_obs_keys = [par["obs_area"], par["obs_aspect_ratio"]]
+    
+    # Add multiple keys
+    for key_list in [par["obs_dapi_stain"], par["obs_membrane_stain"], par["obs_panck_stain"],
+                     par["obs_cd45_stain"], par["obs_cd3_stain"]]:
+                     if key_list:
+                         cosmx_obs_keys.extend(key_list)
+                     
+    # Check if keys are present
+    check_optional_obs_keys(mod_obs, cosmx_obs_keys, "CosMx metrics.")
+    
+    # Filter only existing keys
+    cosmx_formatted_columns = [key for key in cosmx_obs_keys if key in mod_obs]
+    
+    # Add coordinates
+    cosmx_formatted_columns.extend(["x_coord", "y_coord"])
+    
+    for key in cosmx_formatted_columns:
+        if key in mod_obs:
+            if not pd.api.types.is_float_dtype(mod_obs[key]):
+                try:
+                    mod_obs[key] = mod_obs[key].astype("float32")
+                except ValueError:
+                    logger.warning(f"Could not convert column {key} to a float dtype.")
+
+    return mod_obs, cosmx_formatted_columns
+
+
+def generate_cosmx_stats(mod_obs, sample_id, required_keys):
+    
+    # Format required columns
+    mod_obs = format_required_columns(required_keys, mod_obs)
+
+    # Format cosmx-specific columns
+    mod_obs, cosmx_formatted_columns = format_cosmx_columns(mod_obs)
+
+    # Fetch and format  all categorical columns for grouping
+    metadata_obs_keys, mod_obs = format_categorical_columns(mod_obs)
+
+    # Create cell RNA stats dataframe
+    cell_rna_stats = pd.DataFrame(
+        {
+            "sample_id": pd.Categorical(sample_id),
+            **{key: mod_obs[key] for key in required_keys},
+            **{key: mod_obs[key] for key in cosmx_formatted_columns},
+            **{key: mod_obs[key] for key in metadata_obs_keys}
+        }
+    )
+
+    return cell_rna_stats
+
+
+def concatenate_dataframes(dfs):
+    \'\'\'Concatenates a list of dataframes into a single dataframe, preserving categorical columns.\'\'\'
+    df = pd.concat(dfs, ignore_index=True)
+
+    # Find categorical columns that became object columms
+    for col in df.columns:
+        if any(df[col].dtype.name == 'category' for df in dfs if col in df.columns):
+            # Get all categorical series for this column
+            cat_series = [df[col] for df in dfs if col in df.columns and df[col].dtype.name == 'category']
+            if cat_series:
+                # Union the categories and apply to result
+                unioned = pd.api.types.union_categoricals(cat_series)
+                df[col] = pd.Categorical(df[col], categories=unioned.categories)
+    return df
+
+
+def main(par):
+    cell_stats_dfs = []
+    sample_stats_dfs = []
+    metrics_cellranger_dfs = []
+
+    for i, mudata_file in enumerate(par["input"]):
+        logger.info(f"Processing {mudata_file}")
+
+        # read h5mu file
+        file = h5py.File(mudata_file, "r")
+
+        # read the necessary info
+        grp_mod = file["mod"][par["modality"]]
+        mod_obs = ad.experimental.read_elem(grp_mod["obs"])
+        mod_obsm = ad.experimental.read_elem(grp_mod["obsm"])
+        uns = ad.experimental.read_elem(file["uns"])
+
+        # close the h5mu file
+        file.close()
+
+        barcodes_original_count = mod_obs.shape[0]
+
+        # Add coordinates to obs before filtering
+        if par["ingestion_method"] == "xenium" or par["ingestion_method"] == "visium" or par["ingestion_method"] == "cosmx":
+            mod_obs["x_coord"] = mod_obsm["spatial"][:, 0]
+            mod_obs["y_coord"] = mod_obsm["spatial"][:, 1]
+
+        # Pre-filter cells
+        logger.info("Pre-filtering cells based on counts...")
+        if "min_total_counts" in par:
+            mod_obs = mod_obs[mod_obs["total_counts"] >= par["min_total_counts"]]
+        if "min_num_nonzero_vars" in par:
+            mod_obs = mod_obs[mod_obs["num_nonzero_vars"] >= par["min_num_nonzero_vars"]]
+        barcodes_filtered_count = mod_obs.shape[0]
+
+        # Detect sample id's
+        logger.info("Detecting sample id's...")
+        sample_id = (
+            mod_obs[par["obs_sample_id"]].tolist()
+            if par["obs_sample_id"] in mod_obs.columns
+            else [f"sample_{i}"] * mod_obs.shape[0]
+        )
+
+        # Generating sample summary statistics
+        logger.info("Generating sample summary statistics...")
+        required_keys = [
+            par["obs_total_counts"],
+            par["obs_num_nonzero_vars"],
+            par["obs_fraction_mitochondrial"],
+            par["obs_fraction_ribosomal"]
+            ]
+        missing_keys = [key for key in required_keys if key not in mod_obs.columns]
+        if missing_keys:
+            raise ValueError(f"Missing keys in obs: {', '.join(missing_keys)}")
+
+        sample_summary = {
+            "sample_id": pd.Categorical([sample_id[0]]),
+            "rna_num_barcodes": [barcodes_original_count],
+            "rna_num_barcodes_filtered": [barcodes_filtered_count],
+            "rna_sum_total_counts": [mod_obs[par["obs_total_counts"]].sum()],
+            "rna_median_total_counts": [mod_obs[par["obs_total_counts"]].median()],
+            "rna_overall_num_nonzero_vars": [mod_obs[par["obs_num_nonzero_vars"]].sum()],
+            "rna_median_num_nonzero_vars": [mod_obs[par["obs_num_nonzero_vars"]].median()],
+        }
+
+        if par["ingestion_method"] == "xenium":
+            sample_summary["control_probe_percentage"] = mod_obs[par["obs_control_probe_counts"]].sum() / mod_obs["total_counts"].sum() * 100
+            sample_summary["negative_decoding_percentage"] = mod_obs[par["obs_control_codeword_counts"]].sum() / mod_obs["total_counts"].sum() * 100
+
+        sample_summary_stats = pd.DataFrame(sample_summary)
+
+        if par["ingestion_method"] == "cellranger_multi":
+            cell_rna_stats, cellranger_stats = generate_cellranger_stats(mod_obs, uns, sample_id, required_keys)
+            metrics_cellranger_dfs.append(cellranger_stats)
+
+        if par["ingestion_method"] == "xenium":
+            cell_rna_stats = generate_xenium_stats(mod_obs, sample_id, required_keys)
+        
+        if par["ingestion_method"] == "visium":
+            cell_rna_stats = generate_visium_stats(mod_obs, sample_id, required_keys)
+        
+        if par["ingestion_method"] == "cosmx":
+            cell_rna_stats = generate_cosmx_stats(mod_obs, sample_id, required_keys)
+
+        cell_stats_dfs.append(cell_rna_stats)
+        sample_stats_dfs.append(sample_summary_stats)
+
+    # Combine dataframes of all samples
+    logger.info("Combining data of all samples into single object...")
+    combined_cell_stats = concatenate_dataframes(cell_stats_dfs)
+    combined_sample_stats = concatenate_dataframes(sample_stats_dfs)
+    if par["ingestion_method"] == "cellranger_multi":
+        combined_metrics_cellranger = concatenate_dataframes(metrics_cellranger_dfs)
+
+    report_categories = [combined_cell_stats, combined_sample_stats]
+
+    if par["ingestion_method"] == "cellranger_multi":
+        report_categories.append(combined_metrics_cellranger)
+
+    for df in report_categories:
+        df["sample_id"] = pd.Categorical(df["sample_id"])
+
+    output = {
+        "cell_rna_stats": transform_df(combined_cell_stats),
+        "sample_summary_stats": transform_df(combined_sample_stats)
+    }
+
+    if par["ingestion_method"] == "cellranger_multi":
+        output["metrics_cellranger_stats"] = transform_df(combined_metrics_cellranger)
+
+    logger.info(f"Writing output data json to {par['output']}")
+    output_path = Path(par["output"])
+    with open(output_path, "w") as f:
+        json.dump(output, f, indent=2)
+
+    report_structures = {
+        "cellranger_multi": os.path.join(meta["resources_dir"], "report_structure/cellranger.json"),
+        "xenium": os.path.join(meta["resources_dir"], "report_structure/xenium.json"),
+        "visium": os.path.join(meta["resources_dir"], "report_structure/visium.json"),
+        "cosmx": os.path.join(meta["resources_dir"], "report_structure/cosmx.json")
+    }
+
+    logger.info(f"Writing output report structure json to {par['output_reporting_json']}")
+    shutil.copy(report_structures[par["ingestion_method"]], par["output_reporting_json"])
+
+
+if __name__ == "__main__":
+    main(par)
+VIASHMAIN
+python -B "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline_qc/ingestion_qc/h5mu_to_qc_json",
+    "tag" : "v0.2.2"
+  },
+  "label" : [
+    "midmem",
+    "middisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/nextflow.config b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/nextflow.config
new file mode 100644
index 0000000..4ee5342
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'ingestion_qc/h5mu_to_qc_json'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v0.2.2'
+  description = 'Takes H5MU files that have been ingested by CellRanger, Xenium or CosMx and processed by the QC workflow, and generates:\n- A JSON file that contains the combined data for the QC report\n- A JSON file that defines the layout and structure of the QC report\n'
+  author = 'Jakub Majercik, Dorien Roosen, Robrecht Cannoodt, Weiwei Schultz'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/nextflow_labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/cellranger.json b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/cellranger.json
new file mode 100644
index 0000000..18769ba
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/cellranger.json
@@ -0,0 +1,162 @@
+{
+  "categories": [
+    {
+      "name": "Sample QC",
+      "key": "sample_summary_stats",
+      "additionalAxes": false,
+      "defaultFilters": []
+    },
+    {
+      "name": "SampleQC",
+      "key": "metrics_cellranger_stats",
+      "additionalAxes": false,
+      "defaultFilters": [
+        {
+          "type": "bar",
+          "field": "Number_of_reads_in_the_library",
+          "label": "Number of reads per library",
+          "description": "Sequencing depth per sample. Higher values generally indicate more comprehensive cell profiling.",
+          "nBins": 10,
+          "groupBy": "sample_id",
+          "xAxisType": "linear",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "bar",
+          "field": "Confidently_mapped_reads_in_cells",
+          "label": "Confidently mapped reads in cells",
+          "description": "Number of reads that were mapped unambiguously to the reference genome within cell-containing droplets.",
+          "groupBy": "sample_id",
+          "nBins": 10,
+          "yAxisType": "linear"
+        },
+        {
+          "type": "bar",
+          "field": "Estimated_number_of_cells",
+          "label": "Estimated number of cells",
+          "description": "CellRanger's estimate of the number of cells per sample based on the UMI count distribution.",
+          "groupBy": "sample_id",
+          "nBins": 10,
+          "yAxisType": "linear"
+        },
+        {
+          "type": "bar",
+          "field": "Sequencing_saturation",
+          "label": "Sequencing saturation",
+          "description": "Fraction of reads that are duplicates of existing UMIs. Higher values suggest deeper sequencing coverage.",
+          "groupBy": "sample_id",
+          "nBins": 10,
+          "yAxisType": "linear"
+        }
+      ]
+    },
+    {
+      "name": "Cell RNA QC",
+      "key": "cell_rna_stats",
+      "additionalAxes": true,
+      "defaultFilters": [
+        {
+          "type": "histogram",
+          "field": "total_counts",
+          "label": "Total UMI per cell",
+          "description": "Total number of RNA molecules detected per cell. Low values typically indicate empty droplets or low-quality cells that should be filtered out.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "num_nonzero_vars",
+          "label": "Number of non-zero genes per cell",
+          "description": "Count of unique genes detected in each cell. Low gene counts often indicate poor-quality cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "fraction_mitochondrial",
+          "label": "Fraction UMI of mitochondrial genes per cell",
+          "description": "Proportion of cell's RNA from mitochondrial genes.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "fraction_ribosomal",
+          "label": "Fraction UMI of ribosomal genes per cell",
+          "description": "Proportion of cell's RNA from ribosomal protein genes. Extreme values may indicate stress responses or cell cycle abnormalities.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "pct_of_counts_in_top_50_vars",
+          "label": "Fraction UMI in top 50 genes per cell",
+          "description": "Proportion of RNA molecules from the 50 most-expressed genes in each cell.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_cell_probability",
+          "label": "CellBender cell probability",
+          "description": "CellBender's statistical confidence (0-1) that a barcode represents a real cell, with higher values indicating stronger confidence.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_background_fraction",
+          "label": "CellBender background fraction",
+          "description": "Estimated percentage of each cell's RNA that comes from the ambient solution rather than the cell itself.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_cell_size",
+          "label": "CellBender cell size",
+          "description": "CellBender's estimate of the true number of RNA molecules in each cell after removing ambient contamination. Reflects actual cell RNA content rather than raw UMI counts.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "field": "cellbender_droplet_efficiency",
+          "label": "CellBender droplet efficiency",
+          "description": "CellBender's estimate of how efficiently each droplet captured RNA molecules. Higher values indicate more reliable RNA sampling within individual droplets.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        }
+      ]
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/cosmx.json b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/cosmx.json
new file mode 100644
index 0000000..108961e
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/cosmx.json
@@ -0,0 +1,211 @@
+{
+  "categories": [
+    {
+      "name": "Sample QC",
+      "key": "sample_summary_stats",
+      "additionalAxes": false,
+      "defaultFilters": []
+    },
+    {
+      "name": "Cell RNA QC",
+      "key": "cell_rna_stats",
+      "additionalAxes": true,
+      "defaultFilters": [
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "total_counts",
+          "label": "Total UMI per cell",
+          "description": "Total number of RNA molecules detected per cell. Low values typically indicate empty droplets or low-quality cells that should be filtered out.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "num_nonzero_vars",
+          "label": "Number of non-zero genes per cell",
+          "description": "Count of unique genes detected in each cell. Low gene counts often indicate poor-quality cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "fraction_mitochondrial",
+          "label": "Fraction UMI of mitochondrial genes per cell",
+          "description": "Proportion of cell's RNA from mitochondrial genes.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "fraction_ribosomal",
+          "label": "Fraction UMI of ribosomal genes per cell",
+          "description": "Proportion of cell's RNA from ribosomal protein genes. Extreme values may indicate stress responses or cell cycle abnormalities.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "Area",
+          "label": "Cell Area",
+          "description": "Area of the segmented cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "AspectRatio",
+          "label": "Aspect Ratio",
+          "description": "Aspect ratio of the segmented cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "Mean.DAPI",
+          "label": "Mean DAPI Intensity",
+          "description": "Mean intensity of DAPI stain.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "Mean.MembraneStain",
+          "label": "Mean Membrane Stain Intensity",
+          "description": "Mean intensity of Membrane stain.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "Mean.PanCK",
+          "label": "Mean PanCK Intensity",
+          "description": "Mean intensity of PanCK stain.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "Mean.CD45",
+          "label": "Mean CD45 Intensity",
+          "description": "Mean intensity of CD45 stain.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "Mean.CD3",
+          "label": "Mean CD3 Intensity",
+          "description": "Mean intensity of CD3 stain.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "Max.DAPI",
+          "label": "Max DAPI Intensity",
+          "description": "Maximum intensity of DAPI stain.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "Max.MembraneStain",
+          "label": "Max Membrane Stain Intensity",
+          "description": "Maximum intensity of Membrane stain.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "Max.PanCK",
+          "label": "Max PanCK Intensity",
+          "description": "Maximum intensity of PanCK stain.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "Max.CD45",
+          "label": "Max CD45 Intensity",
+          "description": "Maximum intensity of CD45 stain.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "Max.CD3",
+          "label": "Max CD3 Intensity",
+          "description": "Maximum intensity of CD3 stain.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        }
+      ]
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/visium.json b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/visium.json
new file mode 100644
index 0000000..3dae87e
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/visium.json
@@ -0,0 +1,67 @@
+{
+  "categories": [
+    {
+      "name": "Sample QC",
+      "key": "sample_summary_stats",
+      "additionalAxes": false,
+      "defaultFilters": []
+    },
+    {
+      "name": "Cell RNA QC",
+      "key": "cell_rna_stats",
+      "additionalAxes": true,
+      "defaultFilters": [
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "total_counts",
+          "label": "Total UMI per cell",
+          "description": "Total number of RNA molecules detected per cell. Low values typically indicate empty droplets or low-quality cells that should be filtered out.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "num_nonzero_vars",
+          "label": "Number of non-zero genes per cell",
+          "description": "Count of unique genes detected in each cell. Low gene counts often indicate poor-quality cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "fraction_mitochondrial",
+          "label": "Fraction UMI of mitochondrial genes per cell",
+          "description": "Proportion of cell's RNA from mitochondrial genes.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "fraction_ribosomal",
+          "label": "Fraction UMI of ribosomal genes per cell",
+          "description": "Proportion of cell's RNA from ribosomal protein genes. Extreme values may indicate stress responses or cell cycle abnormalities.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        }
+      ]
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/xenium.json b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/xenium.json
new file mode 100644
index 0000000..9dc7a8b
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/report_structure/xenium.json
@@ -0,0 +1,91 @@
+{
+  "categories": [
+    {
+      "name": "Sample QC",
+      "key": "sample_summary_stats",
+      "additionalAxes": false,
+      "defaultFilters": []
+    },
+    {
+      "name": "Cell RNA QC",
+      "key": "cell_rna_stats",
+      "additionalAxes": true,
+      "defaultFilters": [
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "total_counts",
+          "label": "Total UMI per cell",
+          "description": "Total number of RNA molecules detected per cell. Low values typically indicate empty droplets or low-quality cells that should be filtered out.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "num_nonzero_vars",
+          "label": "Number of non-zero genes per cell",
+          "description": "Count of unique genes detected in each cell. Low gene counts often indicate poor-quality cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "zoomMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "fraction_mitochondrial",
+          "label": "Fraction UMI of mitochondrial genes per cell",
+          "description": "Proportion of cell's RNA from mitochondrial genes.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "fraction_ribosomal",
+          "label": "Fraction UMI of ribosomal genes per cell",
+          "description": "Proportion of cell's RNA from ribosomal protein genes. Extreme values may indicate stress responses or cell cycle abnormalities.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "cell_area",
+          "label": "Segmented cell area",
+          "description": "Area of the segmented cells.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        },
+        {
+          "type": "histogram",
+          "visualizationType": "histogram",
+          "field": "nucleus_ratio",
+          "label": "Nucleus Ratio",
+          "description": "Ratio of the nucleus area to the segmented cell area.",
+          "cutoffMin": null,
+          "cutoffMax": null,
+          "nBins": 50,
+          "groupBy": "sample_id",
+          "yAxisType": "linear"
+        }
+      ]
+    }
+  ]
+}
\ No newline at end of file
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/setup_logger.py b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/setup_logger.py
new file mode 100644
index 0000000..3ca1cdb
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/_private/nextflow/ingestion_qc/h5mu_to_qc_json/setup_logger.py
@@ -0,0 +1,12 @@
+def setup_logger():
+    import logging
+    from sys import stdout
+
+    logger = logging.getLogger()
+    logger.setLevel(logging.INFO)
+    console_handler = logging.StreamHandler(stdout)
+    logFormatter = logging.Formatter("%(asctime)s %(levelname)-8s %(message)s")
+    console_handler.setFormatter(logFormatter)
+    logger.addHandler(console_handler)
+
+    return logger
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/.config.vsh.yaml
new file mode 100644
index 0000000..c22f99d
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/.config.vsh.yaml
@@ -0,0 +1,267 @@
+name: "generate_html"
+namespace: "ingestion_qc"
+version: "v0.2.2"
+authors:
+- name: "Jakub Majercik"
+  roles:
+  - "author"
+  info:
+    role: "Contributor"
+    links:
+      email: "jakub@data-intuitive.com"
+      github: "jakubmajercik"
+      linkedin: "jakubmajercik"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Bioinformatics Engineer"
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "file"
+    name: "--input_data"
+    description: "The input JSON file containing the QC metrics"
+    info: null
+    example:
+    - "path/to/file.json"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input_structure"
+    description: "The input JSON file containing the structure of the data"
+    info: null
+    example:
+    - "path/to/file.json"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output_qc_report"
+    description: "The output HTML report"
+    info: null
+    example:
+    - "path/to/file.html"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "bash_script"
+  path: "script.sh"
+  is_executable: true
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Generate an HTML report from the QC metrics"
+test_resources:
+- type: "bash_script"
+  path: "test.sh"
+  is_executable: true
+- type: "file"
+  path: "sc_dataset.json"
+- type: "file"
+  path: "sc_report_structure.json"
+- type: "file"
+  path: "xenium_dataset.json"
+- type: "file"
+  path: "xenium_report_structure.json"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+repositories:
+- type: "vsh"
+  name: "openpipeline"
+  repo: "openpipeline"
+  tag: "v4.0.0"
+- type: "vsh"
+  name: "craftbox"
+  repo: "craftbox"
+  tag: "v0.2.0"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline_qc"
+  docker_registry: "ghcr.io"
+runners:
+- type: "executable"
+  id: "executable"
+  docker_setup_strategy: "ifneedbepullelsecachedbuild"
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    label:
+    - "lowmem"
+    - "lowdisk"
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "docker"
+  id: "docker"
+  image: "node:latest"
+  target_registry: "images.viash-hub.com"
+  target_tag: "v0.2.2"
+  namespace_separator: "/"
+  setup:
+  - type: "apt"
+    packages:
+    - "git"
+    interactive: false
+  - type: "docker"
+    run:
+    - "cd /opt && git clone -b v0.2.1 https://github.com/openpipelines-bio/siqc.git\
+      \ \\\n&& cd siqc && npm install \\\n&& true\n"
+  entrypoint: []
+  cmd: null
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/ingestion_qc/generate_html/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "docker|native"
+  output: "target/nextflow/ingestion_qc/generate_html"
+  executable: "target/nextflow/ingestion_qc/generate_html/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "9b81d310d031b120755e39001cc40f8ec51bc7c2"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline_qc"
+package_config:
+  name: "openpipeline_qc"
+  version: "v0.2.2"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-bio/openpipeline_incubator/resources_test"
+      dest: "resources_test"
+  repositories:
+  - type: "vsh"
+    name: "openpipeline"
+    repo: "openpipeline"
+    tag: "v4.0.0"
+  - type: "vsh"
+    name: "craftbox"
+    repo: "craftbox"
+    tag: "v0.2.0"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag\
+    \ := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v0.2.2'"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline_qc"
+    docker_registry: "ghcr.io"
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/main.nf b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/main.nf
new file mode 100644
index 0000000..25c856a
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/main.nf
@@ -0,0 +1,3956 @@
+// generate_html v0.2.2
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Jakub Majercik (author)
+//  * Dorien Roosen (author)
+//  * Robrecht Cannoodt (author)
+//  * Weiwei Schultz (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "generate_html",
+  "namespace" : "ingestion_qc",
+  "version" : "v0.2.2",
+  "authors" : [
+    {
+      "name" : "Jakub Majercik",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "links" : {
+          "email" : "jakub@data-intuitive.com",
+          "github" : "jakubmajercik",
+          "linkedin" : "jakubmajercik"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Bioinformatics Engineer"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input_data",
+          "description" : "The input JSON file containing the QC metrics",
+          "example" : [
+            "path/to/file.json"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input_structure",
+          "description" : "The input JSON file containing the structure of the data",
+          "example" : [
+            "path/to/file.json"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output_qc_report",
+          "description" : "The output HTML report",
+          "example" : [
+            "path/to/file.html"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "bash_script",
+      "path" : "script.sh",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/src/configs/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Generate an HTML report from the QC metrics",
+  "test_resources" : [
+    {
+      "type" : "bash_script",
+      "path" : "test.sh",
+      "is_executable" : true
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/qc_sample_data/sc_dataset.json"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/qc_sample_data/sc_report_structure.json"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/spatial_qc_sample_data/xenium_dataset.json"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/spatial_qc_sample_data/xenium_report_structure.json"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "repositories" : [
+    {
+      "type" : "vsh",
+      "name" : "openpipeline",
+      "repo" : "openpipeline",
+      "tag" : "v4.0.0"
+    },
+    {
+      "type" : "vsh",
+      "name" : "craftbox",
+      "repo" : "craftbox",
+      "tag" : "v0.2.0"
+    }
+  ],
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline_qc",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "executable",
+      "id" : "executable",
+      "docker_setup_strategy" : "ifneedbepullelsecachedbuild"
+    },
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "label" : [
+          "lowmem",
+          "lowdisk"
+        ],
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "docker",
+      "id" : "docker",
+      "image" : "node:latest",
+      "target_registry" : "images.viash-hub.com",
+      "target_tag" : "v0.2.2",
+      "namespace_separator" : "/",
+      "setup" : [
+        {
+          "type" : "apt",
+          "packages" : [
+            "git"
+          ],
+          "interactive" : false
+        },
+        {
+          "type" : "docker",
+          "run" : [
+            "cd /opt && git clone -b v0.2.1 https://github.com/openpipelines-bio/siqc.git \\\\\n&& cd siqc && npm install \\\\\n&& true\n"
+          ]
+        }
+      ]
+    },
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/ingestion_qc/generate_html/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "docker|native",
+    "output" : "/workdir/root/repo/target/nextflow/ingestion_qc/generate_html",
+    "viash_version" : "0.9.4",
+    "git_commit" : "9b81d310d031b120755e39001cc40f8ec51bc7c2",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline_qc"
+  },
+  "package_config" : {
+    "name" : "openpipeline_qc",
+    "version" : "v0.2.2",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-bio/openpipeline_incubator/resources_test",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "repositories" : [
+      {
+        "type" : "vsh",
+        "name" : "openpipeline",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.0"
+      },
+      {
+        "type" : "vsh",
+        "name" : "craftbox",
+        "repo" : "craftbox",
+        "tag" : "v0.2.0"
+      }
+    ],
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v0.2.2'"
+    ],
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline_qc",
+      "docker_registry" : "ghcr.io"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+
+
+// inner workflow
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  def rawScript = '''set -e
+tempscript=".viash_script.sh"
+cat > "$tempscript" << VIASHMAIN
+## VIASH START
+# The following code has been auto-generated by Viash.
+$( if [ ! -z ${VIASH_PAR_INPUT_DATA+x} ]; then echo "${VIASH_PAR_INPUT_DATA}" | sed "s#'#'\\"'\\"'#g;s#.*#par_input_data='&'#" ; else echo "# par_input_data="; fi )
+$( if [ ! -z ${VIASH_PAR_INPUT_STRUCTURE+x} ]; then echo "${VIASH_PAR_INPUT_STRUCTURE}" | sed "s#'#'\\"'\\"'#g;s#.*#par_input_structure='&'#" ; else echo "# par_input_structure="; fi )
+$( if [ ! -z ${VIASH_PAR_OUTPUT_QC_REPORT+x} ]; then echo "${VIASH_PAR_OUTPUT_QC_REPORT}" | sed "s#'#'\\"'\\"'#g;s#.*#par_output_qc_report='&'#" ; else echo "# par_output_qc_report="; fi )
+$( if [ ! -z ${VIASH_META_NAME+x} ]; then echo "${VIASH_META_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_name='&'#" ; else echo "# meta_name="; fi )
+$( if [ ! -z ${VIASH_META_FUNCTIONALITY_NAME+x} ]; then echo "${VIASH_META_FUNCTIONALITY_NAME}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_functionality_name='&'#" ; else echo "# meta_functionality_name="; fi )
+$( if [ ! -z ${VIASH_META_RESOURCES_DIR+x} ]; then echo "${VIASH_META_RESOURCES_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_resources_dir='&'#" ; else echo "# meta_resources_dir="; fi )
+$( if [ ! -z ${VIASH_META_EXECUTABLE+x} ]; then echo "${VIASH_META_EXECUTABLE}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_executable='&'#" ; else echo "# meta_executable="; fi )
+$( if [ ! -z ${VIASH_META_CONFIG+x} ]; then echo "${VIASH_META_CONFIG}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_config='&'#" ; else echo "# meta_config="; fi )
+$( if [ ! -z ${VIASH_META_TEMP_DIR+x} ]; then echo "${VIASH_META_TEMP_DIR}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_temp_dir='&'#" ; else echo "# meta_temp_dir="; fi )
+$( if [ ! -z ${VIASH_META_CPUS+x} ]; then echo "${VIASH_META_CPUS}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_cpus='&'#" ; else echo "# meta_cpus="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_B+x} ]; then echo "${VIASH_META_MEMORY_B}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_b='&'#" ; else echo "# meta_memory_b="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KB+x} ]; then echo "${VIASH_META_MEMORY_KB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_kb='&'#" ; else echo "# meta_memory_kb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MB+x} ]; then echo "${VIASH_META_MEMORY_MB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_mb='&'#" ; else echo "# meta_memory_mb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GB+x} ]; then echo "${VIASH_META_MEMORY_GB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_gb='&'#" ; else echo "# meta_memory_gb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TB+x} ]; then echo "${VIASH_META_MEMORY_TB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_tb='&'#" ; else echo "# meta_memory_tb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PB+x} ]; then echo "${VIASH_META_MEMORY_PB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_pb='&'#" ; else echo "# meta_memory_pb="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_KIB+x} ]; then echo "${VIASH_META_MEMORY_KIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_kib='&'#" ; else echo "# meta_memory_kib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_MIB+x} ]; then echo "${VIASH_META_MEMORY_MIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_mib='&'#" ; else echo "# meta_memory_mib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_GIB+x} ]; then echo "${VIASH_META_MEMORY_GIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_gib='&'#" ; else echo "# meta_memory_gib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_TIB+x} ]; then echo "${VIASH_META_MEMORY_TIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_tib='&'#" ; else echo "# meta_memory_tib="; fi )
+$( if [ ! -z ${VIASH_META_MEMORY_PIB+x} ]; then echo "${VIASH_META_MEMORY_PIB}" | sed "s#'#'\\"'\\"'#g;s#.*#meta_memory_pib='&'#" ; else echo "# meta_memory_pib="; fi )
+
+## VIASH END
+ABSOLUTE_INPUT_DATA=\\$(realpath \\$par_input_data)
+ABSOLUTE_INPUT_STRUCTURE=\\$(realpath \\$par_input_structure)
+ABSOLUTE_OUTPUT=\\$(realpath \\$par_output_qc_report)
+
+cd /opt/siqc
+mkdir src/data
+
+npm run cli render -- --data "\\$ABSOLUTE_INPUT_DATA" --structure "\\$ABSOLUTE_INPUT_STRUCTURE" --output "\\$ABSOLUTE_OUTPUT"
+VIASHMAIN
+bash "$tempscript"
+'''
+  
+  return vdsl3WorkflowFactory(args, meta, rawScript)
+}
+
+
+
+/**
+  * Generate a workflow for VDSL3 modules.
+  * 
+  * This function is called by the workflowFactory() function.
+  * 
+  * Input channel: [id, input_map]
+  * Output channel: [id, output_map]
+  * 
+  * Internally, this workflow will convert the input channel
+  * to a format which the Nextflow module will be able to handle.
+  */
+def vdsl3WorkflowFactory(Map args, Map meta, String rawScript) {
+  def key = args["key"]
+  def processObj = null
+
+  workflow processWf {
+    take: input_
+    main:
+
+    if (processObj == null) {
+      processObj = _vdsl3ProcessFactory(args, meta, rawScript)
+    }
+    
+    output_ = input_
+      | map { tuple ->
+        def id = tuple[0]
+        def data_ = tuple[1]
+
+        if (workflow.stubRun) {
+          // add id if missing
+          data_ = [id: 'stub'] + data_
+        }
+
+        // process input files separately
+        def inputPaths = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "input" }
+          .collect { par ->
+            def val = data_.containsKey(par.plainName) ? data_[par.plainName] : []
+            def inputFiles = []
+            if (val == null) {
+              inputFiles = []
+            } else if (val instanceof List) {
+              inputFiles = val
+            } else if (val instanceof Path) {
+              inputFiles = [ val ]
+            } else {
+              inputFiles = []
+            }
+            if (!workflow.stubRun) {
+              // throw error when an input file doesn't exist
+              inputFiles.each{ file -> 
+                assert file.exists() :
+                  "Error in module '${key}' id '${id}' argument '${par.plainName}'.\n" +
+                  "  Required input file does not exist.\n" +
+                  "  Path: '$file'.\n" +
+                  "  Expected input file to exist"
+              }
+            }
+            inputFiles 
+          } 
+
+        // remove input files
+        def argsExclInputFiles = meta.config.allArguments
+          .findAll { (it.type != "file" || it.direction != "input") && data_.containsKey(it.plainName) }
+          .collectEntries { par ->
+            def parName = par.plainName
+            def val = data_[parName]
+            if (par.multiple && val instanceof Collection) {
+              val = val.join(par.multiple_sep)
+            }
+            if (par.direction == "output" && par.type == "file") {
+              val = val
+                .replaceAll('\\$id', id)
+                .replaceAll('\\$\\{id\\}', id)
+                .replaceAll('\\$key', key)
+                .replaceAll('\\$\\{key\\}', key)
+            }
+            [parName, val]
+          }
+
+        [ id ] + inputPaths + [ argsExclInputFiles, meta.resources_dir ]
+      }
+      | processObj
+      | map { output ->
+        def outputFiles = meta.config.allArguments
+          .findAll { it.type == "file" && it.direction == "output" }
+          .indexed()
+          .collectEntries{ index, par ->
+            def out = output[index + 1]
+            // strip dummy '.exitcode' file from output (see nextflow-io/nextflow#2678)
+            if (!out instanceof List || out.size() <= 1) {
+              if (par.multiple) {
+                out = []
+              } else {
+                assert !par.required :
+                    "Error in module '${key}' id '${output[0]}' argument '${par.plainName}'.\n" +
+                    "  Required output file is missing"
+                out = null
+              }
+            } else if (out.size() == 2 && !par.multiple) {
+              out = out[1]
+            } else {
+              out = out.drop(1)
+            }
+            [ par.plainName, out ]
+          }
+        
+        // drop null outputs
+        outputFiles.removeAll{it.value == null}
+
+        [ output[0], outputFiles ]
+      }
+    emit: output_
+  }
+
+  return processWf
+}
+
+// depends on: session?
+def _vdsl3ProcessFactory(Map workflowArgs, Map meta, String rawScript) {
+  // autodetect process key
+  def wfKey = workflowArgs["key"]
+  def procKeyPrefix = "${wfKey}_process"
+  def scriptMeta = nextflow.script.ScriptMeta.current()
+  def existing = scriptMeta.getProcessNames().findAll{it.startsWith(procKeyPrefix)}
+  def numbers = existing.collect{it.replace(procKeyPrefix, "0").toInteger()}
+  def newNumber = (numbers + [-1]).max() + 1
+
+  def procKey = newNumber == 0 ? procKeyPrefix : "$procKeyPrefix$newNumber"
+
+  if (newNumber > 0) {
+    log.warn "Key for module '${wfKey}' is duplicated.\n",
+      "If you run a component multiple times in the same workflow,\n" +
+      "it's recommended you set a unique key for every call,\n" +
+      "for example: ${wfKey}.run(key: \"foo\")."
+  }
+
+  // subset directives and convert to list of tuples
+  def drctv = workflowArgs.directives
+
+  // TODO: unit test the two commands below
+  // convert publish array into tags
+  def valueToStr = { val ->
+    // ignore closures
+    if (val instanceof CharSequence) {
+      if (!val.matches('^[{].*[}]$')) {
+        '"' + val + '"'
+      } else {
+        val
+      }
+    } else if (val instanceof List) {
+      "[" + val.collect{valueToStr(it)}.join(", ") + "]"
+    } else if (val instanceof Map) {
+      "[" + val.collect{k, v -> k + ": " + valueToStr(v)}.join(", ") + "]"
+    } else {
+      val.inspect()
+    }
+  }
+
+  // multiple entries allowed: label, publishdir
+  def drctvStrs = drctv.collect { key, value ->
+    if (key in ["label", "publishDir"]) {
+      value.collect{ val ->
+        if (val instanceof Map) {
+          "\n$key " + val.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+        } else if (val == null) {
+          ""
+        } else {
+          "\n$key " + valueToStr(val)
+        }
+      }.join()
+    } else if (value instanceof Map) {
+      "\n$key " + value.collect{ k, v -> k + ": " + valueToStr(v) }.join(", ")
+    } else {
+      "\n$key " + valueToStr(value)
+    }
+  }.join()
+
+  def inputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "input" }
+    .collect { ', path(viash_par_' + it.plainName + ', stageAs: "_viash_par/' + it.plainName + '_?/*")' }
+    .join()
+
+  def outputPaths = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par ->
+      // insert dummy into every output (see nextflow-io/nextflow#2678)
+      if (!par.multiple) {
+        ', path{[".exitcode", args.' + par.plainName + ']}'
+      } else {
+        ', path{[".exitcode"] + args.' + par.plainName + '}'
+      }
+    }
+    .join()
+
+  // TODO: move this functionality somewhere else?
+  if (workflowArgs.auto.transcript) {
+    outputPaths = outputPaths + ', path{[".exitcode", ".command*"]}'
+  } else {
+    outputPaths = outputPaths + ', path{[".exitcode"]}'
+  }
+
+  // create dirs for output files (based on BashWrapper.createParentFiles)
+  def createParentStr = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" && it.create_parent }
+    .collect { par -> 
+      def contents = "args[\"${par.plainName}\"] instanceof List ? args[\"${par.plainName}\"].join('\" \"') : args[\"${par.plainName}\"]"
+      "\${ args.containsKey(\"${par.plainName}\") ? \"mkdir_parent '\" + escapeText(${contents}) + \"'\" : \"\" }"
+    }
+    .join("\n")
+
+  // construct inputFileExports
+  def inputFileExports = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction.toLowerCase() == "input" }
+    .collect { par ->
+      def contents = "viash_par_${par.plainName} instanceof List ? viash_par_${par.plainName}.join(\"${par.multiple_sep}\") : viash_par_${par.plainName}"
+      "\n\${viash_par_${par.plainName}.empty ? \"\" : \"export VIASH_PAR_${par.plainName.toUpperCase()}='\" + escapeText(${contents}) + \"'\"}"
+    }
+
+  // NOTE: if using docker, use /tmp instead of tmpDir!
+  def tmpDir = java.nio.file.Paths.get(
+    System.getenv('NXF_TEMP') ?: 
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('VIASH_TMPDIR') ?: 
+    System.getenv('VIASH_TEMPDIR') ?: 
+    System.getenv('VIASH_TMP') ?: 
+    System.getenv('TEMP') ?: 
+    System.getenv('TMPDIR') ?: 
+    System.getenv('TEMPDIR') ?:
+    System.getenv('TMP') ?: 
+    '/tmp'
+  ).toAbsolutePath()
+
+  // construct stub
+  def stub = meta.config.allArguments
+    .findAll { it.type == "file" && it.direction == "output" }
+    .collect { par -> 
+      "\${ args.containsKey(\"${par.plainName}\") ? \"touch2 \\\"\" + (args[\"${par.plainName}\"] instanceof String ? args[\"${par.plainName}\"].replace(\"_*\", \"_0\") : args[\"${par.plainName}\"].join('\" \"')) + \"\\\"\" : \"\" }"
+    }
+    .join("\n")
+
+  // escape script
+  def escapedScript = rawScript.replace('\\', '\\\\').replace('$', '\\$').replace('"""', '\\"\\"\\"')
+
+  // publishdir assert
+  def assertStr = (workflowArgs.auto.publish == true) || workflowArgs.auto.transcript ? 
+    """\nassert task.publishDir.size() > 0: "if auto.publish is true, params.publish_dir needs to be defined.\\n  Example: --publish_dir './output/'" """ :
+    ""
+
+  // generate process string
+  def procStr = 
+  """nextflow.enable.dsl=2
+  |
+  |def escapeText = { s -> s.toString().replaceAll("'", "'\\\"'\\\"'") }
+  |process $procKey {$drctvStrs
+  |input:
+  |  tuple val(id)$inputPaths, val(args), path(resourcesDir, stageAs: ".viash_meta_resources")
+  |output:
+  |  tuple val("\$id")$outputPaths, optional: true
+  |stub:
+  |\"\"\"
+  |touch2() { mkdir -p "\\\$(dirname "\\\$1")" && touch "\\\$1" ; }
+  |$stub
+  |\"\"\"
+  |script:$assertStr
+  |def parInject = args
+  |  .findAll{key, value -> value != null}
+  |  .collect{key, value -> "export VIASH_PAR_\${key.toUpperCase()}='\${escapeText(value)}'"}
+  |  .join("\\n")
+  |\"\"\"
+  |# meta exports
+  |export VIASH_META_RESOURCES_DIR="\${resourcesDir}"
+  |export VIASH_META_TEMP_DIR="${['docker', 'podman', 'charliecloud'].any{ it == workflow.containerEngine } ? '/tmp' : tmpDir}"
+  |export VIASH_META_NAME="${meta.config.name}"
+  |# export VIASH_META_EXECUTABLE="\\\$VIASH_META_RESOURCES_DIR/\\\$VIASH_META_NAME"
+  |export VIASH_META_CONFIG="\\\$VIASH_META_RESOURCES_DIR/.config.vsh.yaml"
+  |\${task.cpus ? "export VIASH_META_CPUS=\$task.cpus" : "" }
+  |\${task.memory?.bytes != null ? "export VIASH_META_MEMORY_B=\$task.memory.bytes" : "" }
+  |if [ ! -z \\\${VIASH_META_MEMORY_B+x} ]; then
+  |  export VIASH_META_MEMORY_KB=\\\$(( (\\\$VIASH_META_MEMORY_B+999) / 1000 ))
+  |  export VIASH_META_MEMORY_MB=\\\$(( (\\\$VIASH_META_MEMORY_KB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_GB=\\\$(( (\\\$VIASH_META_MEMORY_MB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_TB=\\\$(( (\\\$VIASH_META_MEMORY_GB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_PB=\\\$(( (\\\$VIASH_META_MEMORY_TB+999) / 1000 ))
+  |  export VIASH_META_MEMORY_KIB=\\\$(( (\\\$VIASH_META_MEMORY_B+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_MIB=\\\$(( (\\\$VIASH_META_MEMORY_KIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_GIB=\\\$(( (\\\$VIASH_META_MEMORY_MIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_TIB=\\\$(( (\\\$VIASH_META_MEMORY_GIB+1023) / 1024 ))
+  |  export VIASH_META_MEMORY_PIB=\\\$(( (\\\$VIASH_META_MEMORY_TIB+1023) / 1024 ))
+  |fi
+  |
+  |# meta synonyms
+  |export VIASH_TEMP="\\\$VIASH_META_TEMP_DIR"
+  |export TEMP_DIR="\\\$VIASH_META_TEMP_DIR"
+  |
+  |# create output dirs if need be
+  |function mkdir_parent {
+  |  for file in "\\\$@"; do 
+  |    mkdir -p "\\\$(dirname "\\\$file")"
+  |  done
+  |}
+  |$createParentStr
+  |
+  |# argument exports${inputFileExports.join()}
+  |\$parInject
+  |
+  |# process script
+  |${escapedScript}
+  |\"\"\"
+  |}
+  |""".stripMargin()
+
+  // TODO: print on debug
+  // if (workflowArgs.debug == true) {
+  //   println("######################\n$procStr\n######################")
+  // }
+
+  // write process to temp file
+  def tempFile = java.nio.file.Files.createTempFile("viash-process-${procKey}-", ".nf")
+  addShutdownHook { java.nio.file.Files.deleteIfExists(tempFile) }
+  tempFile.text = procStr
+
+  // create process from temp file
+  def binding = new nextflow.script.ScriptBinding([:])
+  def session = nextflow.Nextflow.getSession()
+  def parser = _getScriptLoader(session)
+    .setModule(true)
+    .setBinding(binding)
+  def moduleScript = parser.runScript(tempFile)
+    .getScript()
+
+  // register module in meta
+  def module = new nextflow.script.IncludeDef.Module(name: procKey)
+  scriptMeta.addModule(moduleScript, module.name, module.alias)
+
+  // retrieve and return process from meta
+  return scriptMeta.getProcess(procKey)
+}
+
+// use Reflection to get a ScriptParser / ScriptLoader
+//   <25.02.0-edge: new nextflow.script.ScriptParser(session)
+//   >=25.02.0-edge: nextflow.script.ScriptLoaderFactory.create(session)
+def _getScriptLoader(nextflow.Session session) {
+  // try using the old method
+  try {
+    Class<?> scriptParserClass = Class.forName('nextflow.script.ScriptParser')
+    return scriptParserClass.getDeclaredConstructor(nextflow.Session).newInstance(session)
+  } catch (ClassNotFoundException e) {
+    // else try with the new method
+    try {
+      Class<?> scriptLoaderFactoryClass = Class.forName('nextflow.script.ScriptLoaderFactory')
+      def createMethod = scriptLoaderFactoryClass.getDeclaredMethod('create', nextflow.Session)
+      return createMethod.invoke(null, session) // null because create is static
+    } catch (ClassNotFoundException | NoSuchMethodException | IllegalAccessException | java.lang.reflect.InvocationTargetException e2) {
+      // Handle the case where neither class is found
+      throw new Exception("Neither nextflow.script.ScriptParser nor nextflow.script.ScriptLoaderFactory could be found. Is this a compatible Nextflow version?", e2)
+    }
+  }
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "container" : {
+    "registry" : "images.viash-hub.com",
+    "image" : "vsh/openpipeline_qc/ingestion_qc/generate_html",
+    "tag" : "v0.2.2"
+  },
+  "label" : [
+    "lowmem",
+    "lowdisk"
+  ],
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/nextflow.config b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/nextflow.config
new file mode 100644
index 0000000..e0b250c
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'ingestion_qc/generate_html'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v0.2.2'
+  description = 'Generate an HTML report from the QC metrics'
+  author = 'Jakub Majercik, Dorien Roosen, Robrecht Cannoodt, Weiwei Schultz'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/nextflow_labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/nextflow_schema.json
new file mode 100644
index 0000000..e21af43
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/ingestion_qc/generate_html/nextflow_schema.json
@@ -0,0 +1,66 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "generate_html",
+  "description": "Generate an HTML report from the QC metrics",
+  "type": "object",
+  "$defs": {
+    "inputs": {
+      "title": "Inputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input_data": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "The input JSON file containing the QC metrics",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"path/to/file.json\"`. "
+        },
+        "input_structure": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "The input JSON file containing the structure of the data",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"path/to/file.json\"`. "
+        }
+      }
+    },
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output_qc_report": {
+          "type": "string",
+          "format": "path",
+          "description": "The output HTML report",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output_qc_report.html\"`, direction: `output`, example: `\"path/to/file.html\"`. ",
+          "default": "$id.$key.output_qc_report.html"
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/inputs"
+    },
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/.config.vsh.yaml b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/.config.vsh.yaml
new file mode 100644
index 0000000..da6321b
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/.config.vsh.yaml
@@ -0,0 +1,508 @@
+name: "generate_qc_report"
+namespace: "workflows"
+version: "v0.2.2"
+authors:
+- name: "Jakub Majercik"
+  roles:
+  - "author"
+  info:
+    role: "Contributor"
+    links:
+      email: "jakub@data-intuitive.com"
+      github: "jakubmajercik"
+      linkedin: "jakubmajercik"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Bioinformatics Engineer"
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Robrecht Cannoodt"
+  roles:
+  - "author"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "robrecht@data-intuitive.com"
+      github: "rcannood"
+      orcid: "0000-0003-3641-729X"
+      linkedin: "robrechtcannoodt"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Science Engineer"
+    - name: "Open Problems"
+      href: "https://openproblems.bio"
+      role: "Core Member"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Inputs"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "The sample IDs to include in the report. If not provided,\nthe sample\
+      \ IDs will be extracted from the h5mu files.\n"
+    info: null
+    example:
+    - "sample1"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    description: "The input h5mu files."
+    info: null
+    example:
+    - "path/to/file1.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ingestion_method"
+    info: null
+    required: true
+    choices:
+    - "cellranger_multi"
+    - "xenium"
+    - "cosmx"
+    - "visium"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--sample_metadata"
+    description: "The sample metadata file corresponding to .obs fields in the h5mu\
+      \ input files, to be used for grouping in the report.\n"
+    info: null
+    example:
+    - "path/to/file.csv"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_samples_per_report"
+    description: "The maximum number of samples to be included per report. \nMultiple\
+      \ reports will be generated (with samples equally divided over all reports)\
+      \ if number of input samples exceeds this threshold.\n"
+    info: null
+    default:
+    - 20
+    required: false
+    min: 1
+    max: 20
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Options"
+  arguments:
+  - type: "string"
+    name: "--var_gene_names"
+    description: "The column name in the .var h5mu files that contains the gene names.\
+      \ If not provided, .var_names will be used.\n"
+    info: null
+    example:
+    - "gene_symbol"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_metadata"
+    description: "The metadata keys in the h5mu .obs to include in the report."
+    info: null
+    example:
+    - "donor_id"
+    - "cell_type"
+    - "batch"
+    - "condition"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "QC options"
+  arguments:
+  - type: "string"
+    name: "--var_name_mitochondrial_genes"
+    description: "In which .var slot to store a boolean array corresponding the mitochondrial\
+      \ genes.\n"
+    info: null
+    default:
+    - "mitochondrial"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_ribosomal_genes"
+    description: "In which .var slot to store a boolean array corresponding the ribosomal\
+      \ genes.\n"
+    info: null
+    default:
+    - "ribosomal"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_total_counts"
+    description: "Minimum total counts for a cell to be included in the output.\n"
+    info: null
+    default:
+    - 10
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_num_nonzero_vars"
+    description: "Minimum number of nonzero vars for a cell to be included in the\
+      \ output.\n"
+    info: null
+    default:
+    - 10
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Cellbender options"
+  arguments:
+  - type: "boolean"
+    name: "--run_cellbender"
+    description: "Whether to run cellbender or not."
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--cellbender_epochs"
+    description: "Number of epochs to train cellbender."
+    info: null
+    default:
+    - 150
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Options for CosMx reports"
+  arguments:
+  - type: "string"
+    name: "--obs_area"
+    description: "The key in the h5mu .obs field that contains the cell area."
+    info: null
+    default:
+    - "Area"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_aspect_ratio"
+    description: "The key in the h5mu .obs field that contains the aspect ratio."
+    info: null
+    default:
+    - "AspectRatio"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_dapi_stain"
+    description: "The keys in the h5mu .obs field that contains the DAPI stain intensity."
+    info: null
+    default:
+    - "Mean.DAPI"
+    - "Max.DAPI"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_membrane_stain"
+    description: "The keys in the h5mu .obs field that contains the membrane stain\
+      \ intensity."
+    info: null
+    default:
+    - "Mean.MembraneStain"
+    - "Max.MembraneStain"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_panck_stain"
+    description: "The keys in the h5mu .obs field that contains the PanCK stain intensity."
+    info: null
+    default:
+    - "Mean.PanCK"
+    - "Max.PanCK"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_cd45_stain"
+    description: "The keys in the h5mu .obs field that contains the CD45 stain intensity."
+    info: null
+    default:
+    - "Mean.CD45"
+    - "Max.CD45"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_cd3_stain"
+    description: "The keys in the h5mu .obs field that contains the CD3 stain intensity."
+    info: null
+    default:
+    - "Mean.CD3"
+    - "Max.CD3"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output_qc_report"
+    description: "The output HTML report"
+    info: null
+    example:
+    - "path/to/file.html"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_processed_h5mu"
+    description: "Folder containing the processed h5mu files."
+    info: null
+    default:
+    - "qc_h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "Run the ingestion QC report generation"
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_no_cellbender"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_with_cellbender"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_xenium"
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_visium"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+dependencies:
+- name: "metadata/add_id"
+  repository:
+    type: "vsh"
+    repo: "openpipeline"
+    tag: "v4.0.0"
+- name: "workflows/qc/qc"
+  repository:
+    type: "vsh"
+    repo: "openpipeline"
+    tag: "v4.0.0"
+- name: "correction/cellbender_remove_background"
+  alias: "cellbender"
+  repository:
+    type: "vsh"
+    repo: "openpipeline"
+    tag: "v4.0.0"
+- name: "ingestion_qc/h5mu_to_qc_json"
+  repository:
+    type: "local"
+- name: "ingestion_qc/generate_html"
+  repository:
+    type: "local"
+- name: "move_files_to_directory"
+  repository:
+    type: "vsh"
+    repo: "craftbox"
+    tag: "v0.2.0"
+repositories:
+- type: "vsh"
+  name: "openpipeline"
+  repo: "openpipeline"
+  tag: "v4.0.0"
+- type: "vsh"
+  name: "craftbox"
+  repo: "craftbox"
+  tag: "v0.2.0"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline_qc"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/workflows/generate_qc_report/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/workflows/generate_qc_report"
+  executable: "target/nextflow/workflows/generate_qc_report/main.nf"
+  viash_version: "0.9.4"
+  git_commit: "9b81d310d031b120755e39001cc40f8ec51bc7c2"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline_qc"
+  dependencies:
+  - "target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id"
+  - "target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc"
+  - "target/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background"
+  - "target/_private/nextflow/ingestion_qc/h5mu_to_qc_json"
+  - "target/nextflow/ingestion_qc/generate_html"
+  - "target/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory"
+package_config:
+  name: "openpipeline_qc"
+  version: "v0.2.2"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-bio/openpipeline_incubator/resources_test"
+      dest: "resources_test"
+  repositories:
+  - type: "vsh"
+    name: "openpipeline"
+    repo: "openpipeline"
+    tag: "v4.0.0"
+  - type: "vsh"
+    name: "craftbox"
+    repo: "craftbox"
+    tag: "v0.2.0"
+  viash_version: "0.9.4"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag\
+    \ := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'v0.2.2'"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline_qc"
+    docker_registry: "ghcr.io"
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/main.nf b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/main.nf
new file mode 100644
index 0000000..62e8911
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/main.nf
@@ -0,0 +1,4052 @@
+// generate_qc_report v0.2.2
+// 
+// This wrapper script is auto-generated by viash 0.9.4 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Jakub Majercik (author)
+//  * Dorien Roosen (author)
+//  * Robrecht Cannoodt (author)
+//  * Weiwei Schultz (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  def br = java.nio.file.Files.newBufferedReader(inputFile)
+
+  def row = -1
+  def header = null
+  while (br.ready() && header == null) {
+    def line = br.readLine()
+    row++
+    if (!line.startsWith("#")) {
+      header = splitRegex.split(line, -1).collect{field ->
+        m = removeQuote.matcher(field)
+        m.find() ? m.replaceFirst('$1') : field
+      }
+    }
+  }
+  assert header != null: "CSV file should contain a header"
+
+  while (br.ready()) {
+    def line = br.readLine()
+    row++
+    if (line == null) {
+      br.close()
+      break
+    }
+
+    if (!line.startsWith("#")) {
+      def predata = splitRegex.split(line, -1)
+      def data = predata.collect{field ->
+        if (field == "") {
+          return null
+        }
+        def m = removeQuote.matcher(field)
+        if (m.find()) {
+          return m.replaceFirst('$1')
+        } else {
+          return field
+        }
+      }
+      assert header.size() == data.size(): "Row $row should contain the same number as fields as the header"
+      
+      def dataMap = [header, data].transpose().collectEntries().findAll{it.value != null}
+      output.add(dataMap)
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      if (infile.toString() != outfile.toString()) {
+        [
+          "[ -d \"\$(dirname '${outfile.toString()}')\" ] || mkdir -p \"\$(dirname '${outfile.toString()}')\"",
+          "cp -r '${infile.toString()}' '${outfile.toString()}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "generate_qc_report",
+  "namespace" : "workflows",
+  "version" : "v0.2.2",
+  "authors" : [
+    {
+      "name" : "Jakub Majercik",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "links" : {
+          "email" : "jakub@data-intuitive.com",
+          "github" : "jakubmajercik",
+          "linkedin" : "jakubmajercik"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Bioinformatics Engineer"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Robrecht Cannoodt",
+      "roles" : [
+        "author"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "robrecht@data-intuitive.com",
+          "github" : "rcannood",
+          "orcid" : "0000-0003-3641-729X",
+          "linkedin" : "robrechtcannoodt"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Science Engineer"
+          },
+          {
+            "name" : "Open Problems",
+            "href" : "https://openproblems.bio",
+            "role" : "Core Member"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Inputs",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "The sample IDs to include in the report. If not provided,\nthe sample IDs will be extracted from the h5mu files.\n",
+          "example" : [
+            "sample1"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "The input h5mu files.",
+          "example" : [
+            "path/to/file1.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--ingestion_method",
+          "required" : true,
+          "choices" : [
+            "cellranger_multi",
+            "xenium",
+            "cosmx",
+            "visium"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--sample_metadata",
+          "description" : "The sample metadata file corresponding to .obs fields in the h5mu input files, to be used for grouping in the report.\n",
+          "example" : [
+            "path/to/file.csv"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_samples_per_report",
+          "description" : "The maximum number of samples to be included per report. \nMultiple reports will be generated (with samples equally divided over all reports) if number of input samples exceeds this threshold.\n",
+          "default" : [
+            20
+          ],
+          "required" : false,
+          "min" : 1,
+          "max" : 20,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_gene_names",
+          "description" : "The column name in the .var h5mu files that contains the gene names. If not provided, .var_names will be used.\n",
+          "example" : [
+            "gene_symbol"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_metadata",
+          "description" : "The metadata keys in the h5mu .obs to include in the report.",
+          "example" : [
+            "donor_id",
+            "cell_type",
+            "batch",
+            "condition"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "QC options",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_name_mitochondrial_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "default" : [
+            "mitochondrial"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_ribosomal_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "default" : [
+            "ribosomal"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_total_counts",
+          "description" : "Minimum total counts for a cell to be included in the output.\n",
+          "default" : [
+            10
+          ],
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_num_nonzero_vars",
+          "description" : "Minimum number of nonzero vars for a cell to be included in the output.\n",
+          "default" : [
+            10
+          ],
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Cellbender options",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--run_cellbender",
+          "description" : "Whether to run cellbender or not.",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--cellbender_epochs",
+          "description" : "Number of epochs to train cellbender.",
+          "default" : [
+            150
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Options for CosMx reports",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--obs_area",
+          "description" : "The key in the h5mu .obs field that contains the cell area.",
+          "default" : [
+            "Area"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_aspect_ratio",
+          "description" : "The key in the h5mu .obs field that contains the aspect ratio.",
+          "default" : [
+            "AspectRatio"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_dapi_stain",
+          "description" : "The keys in the h5mu .obs field that contains the DAPI stain intensity.",
+          "default" : [
+            "Mean.DAPI",
+            "Max.DAPI"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_membrane_stain",
+          "description" : "The keys in the h5mu .obs field that contains the membrane stain intensity.",
+          "default" : [
+            "Mean.MembraneStain",
+            "Max.MembraneStain"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_panck_stain",
+          "description" : "The keys in the h5mu .obs field that contains the PanCK stain intensity.",
+          "default" : [
+            "Mean.PanCK",
+            "Max.PanCK"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_cd45_stain",
+          "description" : "The keys in the h5mu .obs field that contains the CD45 stain intensity.",
+          "default" : [
+            "Mean.CD45",
+            "Max.CD45"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_cd3_stain",
+          "description" : "The keys in the h5mu .obs field that contains the CD3 stain intensity.",
+          "default" : [
+            "Mean.CD3",
+            "Max.CD3"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output_qc_report",
+          "description" : "The output HTML report",
+          "example" : [
+            "path/to/file.html"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_processed_h5mu",
+          "description" : "Folder containing the processed h5mu files.",
+          "default" : [
+            "qc_h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/configs/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "Run the ingestion QC report generation",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_no_cellbender"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_with_cellbender"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_xenium"
+    },
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_visium"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "dependencies" : [
+    {
+      "name" : "metadata/add_id",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.0"
+      }
+    },
+    {
+      "name" : "workflows/qc/qc",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.0"
+      }
+    },
+    {
+      "name" : "correction/cellbender_remove_background",
+      "alias" : "cellbender",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.0"
+      }
+    },
+    {
+      "name" : "ingestion_qc/h5mu_to_qc_json",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "ingestion_qc/generate_html",
+      "repository" : {
+        "type" : "local"
+      }
+    },
+    {
+      "name" : "move_files_to_directory",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "craftbox",
+        "tag" : "v0.2.0"
+      }
+    }
+  ],
+  "repositories" : [
+    {
+      "type" : "vsh",
+      "name" : "openpipeline",
+      "repo" : "openpipeline",
+      "tag" : "v4.0.0"
+    },
+    {
+      "type" : "vsh",
+      "name" : "craftbox",
+      "repo" : "craftbox",
+      "tag" : "v0.2.0"
+    }
+  ],
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline_qc",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/workflows/generate_qc_report/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/workflows/generate_qc_report",
+    "viash_version" : "0.9.4",
+    "git_commit" : "9b81d310d031b120755e39001cc40f8ec51bc7c2",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline_qc"
+  },
+  "package_config" : {
+    "name" : "openpipeline_qc",
+    "version" : "v0.2.2",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-bio/openpipeline_incubator/resources_test",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "repositories" : [
+      {
+        "type" : "vsh",
+        "name" : "openpipeline",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.0"
+      },
+      {
+        "type" : "vsh",
+        "name" : "craftbox",
+        "repo" : "craftbox",
+        "tag" : "v0.2.0"
+      }
+    ],
+    "viash_version" : "0.9.4",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'v0.2.2'"
+    ],
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline_qc",
+      "docker_registry" : "ghcr.io"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { add_id } from "${meta.root_dir}/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/metadata/add_id/main.nf"
+include { qc } from "${meta.root_dir}/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/workflows/qc/qc/main.nf"
+include { cellbender_remove_background as cellbender_viashalias } from "${meta.root_dir}/dependencies/vsh/vsh/openpipeline/v4.0.0/nextflow/correction/cellbender_remove_background/main.nf"
+cellbender = cellbender_viashalias.run(key: "cellbender")
+include { h5mu_to_qc_json } from "${meta.resources_dir}/../../../_private/nextflow/ingestion_qc/h5mu_to_qc_json/main.nf"
+include { generate_html } from "${meta.resources_dir}/../../../nextflow/ingestion_qc/generate_html/main.nf"
+include { move_files_to_directory } from "${meta.root_dir}/dependencies/vsh/vsh/craftbox/v0.2.0/nextflow/move_files_to_directory/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take: input_ch
+  main:
+  qc_ch = input_ch
+    // store join id
+    | map { id, state ->
+      [id, state + [_meta: [join_id: id]]]
+    }
+
+    // add sample ids to each state
+    | add_id.run(
+      fromState: [
+        input_id: "id", 
+        input: "input"
+      ],
+      args: [
+        obs_output: "sample_id"
+      ],
+      toState: [ "input": "output" ]
+    )
+
+    // run cellbender
+    | cellbender.run(
+      runIf: {id, state -> state.run_cellbender},
+      fromState: [
+        id: "id",
+        input: "input",
+        epochs: "cellbender_epochs",
+      ],
+      args: [
+        obs_background_fraction: "cellbender_background_fraction",
+        obs_cell_probability: "cellbender_cell_probability",
+        obs_droplet_efficiency: "cellbender_droplet_efficiency",
+        obs_cell_size: "cellbender_cell_size",
+      ],
+      toState: ["input": "output"]
+    )
+
+    // run qc on each sample
+    | qc.run(
+      fromState: [
+        id: "id",
+        input: "input",
+        var_gene_names: "var_gene_names"
+      ],
+      args: [
+        var_name_mitochondrial_genes: "mitochondrial",
+        var_name_ribosomal_genes: "ribosomal",
+        output_obs_num_nonzero_vars: "num_nonzero_vars",
+        output_obs_total_counts_vars: "total_counts"
+      ],
+      toState: { id, output, state ->
+        def keysToRemove = ["var_gene_names", "var_name_mitochondrial_genes", "var_name_ribosomal_genes", "run_cellbender", "cellbender_epochs"]
+        def newState = state.findAll{it.key !in keysToRemove}
+        newState + ["input": output.output]
+      }
+    )
+
+    | joinStates { ids, states ->
+      def newId = "qc_data"
+      // gather keys with unique values across states that should be combined
+      def new_state_non_unique_values = [
+        input: states.collect{it.input},
+        join_ids: states.collect{it._meta.join_id},
+        _meta: [join_id: ids[0]]
+      ]
+      // gather keys from different states
+      def all_state_keys = states.inject([].toSet()){ current_keys, state ->
+            def new_keys = current_keys + state.keySet()
+            return new_keys
+        }.minus(["output", "id", "input", "_meta"])
+      // Create the new state from the keys, values should be the same across samples
+      def new_state = all_state_keys.inject([:]){ old_state, argument_name ->
+            argument_values = states.collect{it.get(argument_name)}.unique()
+            assert argument_values.size() == 1, "Arguments should be the same across samples. Argument name: $argument_name, \
+                                                 argument value: $argument_values"
+            // take the unique value from the set (there is only one)
+            def argument_value
+            argument_values.each { argument_value = it }
+            def current_state = old_state + [(argument_name): argument_value]
+            return current_state
+        }
+      def data_state = new_state_non_unique_values + new_state
+      [ newId, data_state ]
+    }
+
+  processed_files_ch = qc_ch
+
+    // move all processed h5mu files to the same folder
+    | move_files_to_directory.run(
+      fromState: [
+        input: "input",
+        output: "output_processed_h5mu"
+      ],
+      toState: [ "output_processed_h5mu": "output" ]
+    )
+    | setState(["output_processed_h5mu"])
+
+  report_ch = qc_ch
+    // group the processed samples to generate one or multiple reports
+    | flatMap { id, state ->
+
+        // calculate number of reports to be generated and number of samples per report
+        def totalInputs = state.input.size()
+        def maxSamplesPerGroup = state.max_samples_per_report
+        def numGroups = Math.max(1, Math.ceil(totalInputs / maxSamplesPerGroup) as Integer)
+        def baseSamplesPerGroup = totalInputs.intdiv(numGroups)
+        def remainder = totalInputs % numGroups
+
+        println "Dividing ${totalInputs} sample(s) over ${numGroups} report(s) (max ${maxSamplesPerGroup} per report)"
+        
+        // sort inputs to make grouping deterministic
+        def inputs = []
+        for (int i = 0; i < state.input.size(); i++) {
+            inputs << [input: state.input[i], _meta: [join_id: state.join_ids[i]]]
+        }
+
+        def sortedInputs = inputs.sort { it._meta.join_id }
+  
+        def groups = []
+        def itemIndex = 0
+
+        // create one channel per report
+        (0..<numGroups).each { groupNum ->
+            def samplesInGroup = baseSamplesPerGroup + (groupNum < remainder ? 1 : 0)
+            def groupItems = sortedInputs[itemIndex..<(itemIndex + samplesInGroup)]
+            
+            def newId = "combined_${groupNum + 1}_of_${numGroups}"
+            def newState = state.clone()  // Copy all the original state
+            
+            // Override the input and _meta with the grouped items
+            newState.input = groupItems.collect { it.input }
+            newState._meta = groupItems[0]._meta
+            
+            println "Group ${groupNum + 1}: ${samplesInGroup} samples - ${newState._meta}"
+            
+            groups << [newId, newState]
+            itemIndex += samplesInGroup
+        }
+        
+        return groups
+    }
+
+    // Set aside output for QC report instructions
+    | map { id, state -> 
+      def new_state = state + ["output_reporting_json": "reporting_json.json"]
+      [id, new_state]
+    }
+
+    // generate qc json
+    | h5mu_to_qc_json.run(
+      fromState: [
+        input: "input",
+        ingestion_method: "ingestion_method",
+        obs_metadata: "obs_metadata",
+        min_total_counts: "min_total_counts",
+        min_num_nonzero_vars: "min_num_nonzero_vars",
+        obs_area: "obs_area",
+        obs_aspect_ratio: "obs_aspect_ratio",
+        obs_dapi_stain: "obs_dapi_stain",
+        obs_membrane_stain: "obs_membrane_stain",
+        obs_panck_stain: "obs_panck_stain",
+        obs_cd45_stain: "obs_cd45_stain",
+        obs_cd3_stain: "obs_cd3_stain",
+      ],
+      args: [
+        obs_sample_id: "sample_id",
+        obs_total_counts: "total_counts",
+        obs_num_nonzero_vars: "num_nonzero_vars",
+        obs_fraction_mitochondrial: "fraction_mitochondrial",
+        obs_fraction_ribosomal: "fraction_ribosomal",
+      ],
+      toState: [
+        output: "output",
+        output_reporting_json: "output_reporting_json"
+      ]
+    )
+
+    // generate html report
+    | generate_html.run(
+      fromState: [ 
+        input_data: "output",
+        input_structure: "output_reporting_json"
+      ],
+      toState: [
+        output_qc_report: "output_qc_report"
+      ]
+    )
+
+    // collect the reports into a single channel
+    | joinStates { ids, states ->
+      def newId = "qc_report"
+      def report_state = [
+        output_qc_report: states.collect{it.output_qc_report},
+        _meta: states[0]._meta
+      ]
+      [ newId, report_state ]
+    }
+    
+  output_ch = report_ch.mix(processed_files_ch)
+
+    | joinStates { ids, states ->
+
+      assert states.size() == 2, "Expected 2 states, but got ${states.size()}"
+      assert ids.contains('qc_report'), "Expected one channel to have the id `qc_report`, but got ${ids}"
+      assert ids.contains('qc_data'), "Expected one channel to have the id `qc_data`, but got ${ids}"
+
+      def newId = "combined"
+      def combined_state = states[0] + states [1]
+
+      [ newId, combined_state ]
+    }
+
+  emit: output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/nextflow.config b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/nextflow.config
new file mode 100644
index 0000000..4481261
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'workflows/generate_qc_report'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'v0.2.2'
+  description = 'Run the ingestion QC report generation'
+  author = 'Jakub Majercik, Dorien Roosen, Robrecht Cannoodt, Weiwei Schultz'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/nextflow_labels.config b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/nextflow_labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/nextflow_schema.json b/target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/nextflow_schema.json
new file mode 100644
index 0000000..e69de29
diff --git a/target/nextflow/single_cell/cellranger_multi_qc/.config.vsh.yaml b/target/nextflow/single_cell/cellranger_multi_qc/.config.vsh.yaml
new file mode 100644
index 0000000..9069502
--- /dev/null
+++ b/target/nextflow/single_cell/cellranger_multi_qc/.config.vsh.yaml
@@ -0,0 +1,910 @@
+name: "cellranger_multi_qc"
+namespace: "single_cell"
+version: "set-public-scope"
+authors:
+- name: "Jakub Majercik"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Contributor"
+    links:
+      email: "jakub@data-intuitive.com"
+      github: "jakubmajercik"
+      linkedin: "jakubmajercik"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Bioinformatics Engineer"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Input files"
+  arguments:
+  - type: "file"
+    name: "--input"
+    description: "The FASTQ files to be analyzed. FASTQ files should conform to the\
+      \ naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane\
+      \ Number]_[Read Type]_001.fastq.gz`\n"
+    info: null
+    example:
+    - "mysample_S1_L001_R1_001.fastq.gz"
+    - "mysample_S1_L001_R2_001.fastq.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Library arguments"
+  arguments:
+  - type: "string"
+    name: "--library_id"
+    description: "The Illumina sample name to analyze. This must exactly match the\
+      \ 'Sample Name'part\nof the FASTQ files specified in the `--input` argument.\n"
+    info: null
+    example:
+    - "mysample1"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--library_type"
+    description: "The underlying feature type of the library.\n"
+    info: null
+    example:
+    - "Gene Expression"
+    required: false
+    choices:
+    - "Gene Expression"
+    - "VDJ"
+    - "VDJ-T"
+    - "VDJ-B"
+    - "VDJ-T-GD"
+    - "Antibody Capture"
+    - "CRISPR Guide Capture"
+    - "Multiplexing Capture"
+    - "Antigen Capture"
+    - "Custom"
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--library_subsample"
+    description: "The rate at which reads from the provided FASTQ files are sampled.\n\
+      Must be strictly greater than 0 and less than or equal to 1.\n"
+    info: null
+    example:
+    - "0.5"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--library_lanes"
+    description: "Lanes associated with this sample. Defaults to using all lanes."
+    info: null
+    example:
+    - "1-4"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--library_chemistry"
+    description: "Only applicable to FRP. Library-specific assay configuration. By\
+      \ default,\nthe assay configuration is detected automatically. Typically, users\
+      \ will\nnot need to specify a chemistry.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Sample parameters"
+  arguments:
+  - type: "string"
+    name: "--sample_ids"
+    alternatives:
+    - "--cell_multiplex_sample_id"
+    description: "A name to identify a multiplexed sample. Must be alphanumeric with\
+      \ hyphens and/or underscores,\nand less than 64 characters. Required for Cell\
+      \ Multiplexing libraries.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--sample_description"
+    alternatives:
+    - "--cell_multiplex_description"
+    description: "A description for the sample."
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--sample_expect_cells"
+    description: "Expected number of recovered cells, used as input to cell calling\
+      \ algorithm.\n"
+    info: null
+    example:
+    - 3000
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--sample_force_cells"
+    description: "Force pipeline to use this number of cells, bypassing cell detection.\n"
+    info: null
+    example:
+    - 3000
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Feature Barcode library specific arguments"
+  arguments:
+  - type: "file"
+    name: "--feature_reference"
+    description: "Path to the Feature reference CSV file, declaring Feature Barcode\
+      \ constructs and associated barcodes.\nRequired only for Antibody Capture or\
+      \ CRISPR Guide Capture libraries.\nSee https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/using/feature-bc-analysis#feature-ref\
+      \ for more information.\"\n"
+    info: null
+    example:
+    - "feature_reference.csv"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--feature_r1_length"
+    description: "Limit the length of the input Read 1 sequence of V(D)J libraries\
+      \ to the first N bases,\nwhere N is the user-supplied value. Note that the length\
+      \ includes the Barcode and UMI\nsequences so do not set this below 26.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--feature_r2_length"
+    description: "Limit the length of the input Read 2 sequence of V(D)J libraries\
+      \ to the first N bases,\nwhere N is a user-supplied value. Trimming occurs before\
+      \ sequencing metrics are computed\nand therefore, limiting the length of Read\
+      \ 2 may affect Q30 scores.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--min_crispr_umi"
+    description: "Set the minimum number of CRISPR guide RNA UMIs required for protospacer\
+      \ detection.\nIf a lower or higher sensitivity is desired for detection, this\
+      \ value can be customized\naccording to specific experimental needs. Applicable\
+      \ only to datasets that include a\nCRISPR Guide Capture library.\n"
+    info: null
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Gene expression arguments"
+  description: "Arguments relevant to the analysis of gene expression data."
+  arguments:
+  - type: "file"
+    name: "--gex_reference"
+    description: "Genome refence index built by Cell Ranger mkref."
+    info: null
+    example:
+    - "reference_genome.tar.gz"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--gex_secondary_analysis"
+    description: "Whether or not to run the secondary analysis e.g. clustering."
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--gex_generate_bam"
+    description: "Whether to generate a BAM file."
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--tenx_cloud_token_path"
+    description: "The 10x Cloud Analysis user token used to enable cell annotation."
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--cell_annotation_model"
+    description: "\"Cell annotation model to use. If auto, uses the default model\
+      \ for the species.\nIf not given, does not run cell annotation.\"\n"
+    info: null
+    required: false
+    choices:
+    - "auto"
+    - "human_pca_v1_beta"
+    - "mouse_pca_v1_beta"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gex_expect_cells"
+    description: "Expected number of recovered cells, used as input to cell calling\
+      \ algorithm.\n"
+    info: null
+    example:
+    - 3000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gex_force_cells"
+    description: "Force pipeline to use this number of cells, bypassing cell detection.\n"
+    info: null
+    example:
+    - 3000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--gex_include_introns"
+    description: "Whether or not to include intronic reads in counts.\nThis option\
+      \ does not apply to Fixed RNA Profiling analysis.\n"
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gex_r1_length"
+    description: "Limit the length of the input Read 1 sequence of V(D)J libraries\
+      \ to the first N bases,\nwhere N is the user-supplied value. Note that the length\
+      \ includes the Barcode and UMI\nsequences so do not set this below 26.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--gex_r2_length"
+    description: "Limit the length of the input Read 2 sequence of V(D)J libraries\
+      \ to the first N bases,\nwhere N is a user-supplied value. Trimming occurs before\
+      \ sequencing metrics are computed\nand therefore, limiting the length of Read\
+      \ 2 may affect Q30 scores.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--gex_chemistry"
+    description: "Assay configuration. Either specify a single value which will be\
+      \ applied to all libraries,\nor a number of values that is equal to the number\
+      \ of libararies. The latter is only applicable\nto only applicable to Fixed\
+      \ RNA Profiling.\n  - auto: Chemistry autodetection (default)\n  - threeprime:\
+      \ Single Cell 3'\n  - SC3Pv1, SC3Pv2, SC3Pv3(-polyA), SC3Pv4(-polyA): Single\
+      \ Cell 3' v1, v2, v3, or v4\n  - SC3Pv3HT(-polyA): Single Cell 3' v3.1 HT\n\
+      \  - SC-FB: Single Cell Antibody-only 3' v2 or 5'\n  - fiveprime: Single Cell\
+      \ 5'\n  - SC5P-PE: Paired-end Single Cell 5'\n  - SC5P-PE-v3: Paired-end Single\
+      \ Cell 5' v3\n  - SC5P-R2: R2-only Single Cell 5'\n  - SC5P-R2-v3: R2-only Single\
+      \ Cell 5' v3\n  - SCP5-PE-v3: Single Cell 5' paired-end v3 (GEM-X)\n  - SC5PHT\
+      \ : Single Cell 5' v2 HT\n  - SFRP: Fixed RNA Profiling (Singleplex)\n  - MFRP:\
+      \ Fixed RNA Profiling (Multiplex, Probe Barcode on R2)\n  - MFRP-R1: Fixed RNA\
+      \ Profiling (Multiplex, Probe Barcode on R1)\n  - MFRP-RNA: Fixed RNA Profiling\
+      \ (Multiplex, RNA, Probe Barcode on R2)\n  - MFRP-Ab: Fixed RNA Profiling (Multiplex,\
+      \ Antibody, Probe Barcode at R2:69)\n  - MFRP-Ab-R2pos50: Fixed RNA Profiling\
+      \ (Multiplex, Antibody, Probe Barcode at R2:50)\n  - MFRP-RNA-R1: Fixed RNA\
+      \ Profiling (Multiplex, RNA, Probe Barcode on R1)\n  - MFRP-Ab-R1: Fixed RNA\
+      \ Profiling (Multiplex, Antibody, Probe Barcode on R1)\n  - ARC-v1 for analyzing\
+      \ the Gene Expression portion of Multiome data. If Cell Ranger auto-detects\
+      \ ARC-v1 chemistry, an error is triggered.\nSee https://kb.10xgenomics.com/hc/en-us/articles/115003764132-How-does-Cell-Ranger-auto-detect-chemistry-\
+      \ for more information.\n"
+    info: null
+    default:
+    - "auto"
+    required: false
+    choices:
+    - "auto"
+    - "threeprime"
+    - "fiveprime"
+    - "SC3Pv1"
+    - "SC3Pv2"
+    - "SC3Pv3"
+    - "SC3Pv3-polyA"
+    - "SC3Pv4"
+    - "SC3Pv4-polyA"
+    - "SC3Pv3LT"
+    - "SC3Pv3HT"
+    - "SC3Pv3HT-polyA"
+    - "SC5P-PE"
+    - "SC5P-PE-v3"
+    - "SC5P-R2"
+    - "SC-FB"
+    - "SC5P-R2-v3"
+    - "SCP5-PE-v3"
+    - "SC5PHT"
+    - "MFRP"
+    - "MFRP-R1"
+    - "MFRP-RNA"
+    - "MFRP-Ab"
+    - "SFRP"
+    - "MFRP-Ab-R2pos50"
+    - "MFRP-RNA-R1"
+    - "MFRP-Ab-R1"
+    - "ARC-v1"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "VDJ related parameters"
+  arguments:
+  - type: "file"
+    name: "--vdj_reference"
+    description: "VDJ refence index built by Cell Ranger mkref."
+    info: null
+    example:
+    - "reference_vdj.tar.gz"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--vdj_inner_enrichment_primers"
+    description: "V(D)J Immune Profiling libraries: if inner enrichment primers other\
+      \ than those provided \nin the 10x Genomics kits are used, they need to be specified\
+      \ here as a\ntext file with one primer per line.\n"
+    info: null
+    example:
+    - "enrichment_primers.txt"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--vdj_r1_length"
+    description: "Limit the length of the input Read 1 sequence of V(D)J libraries\
+      \ to the first N bases, where N is the user-supplied value.\nNote that the length\
+      \ includes the Barcode and UMI sequences so do not set this below 26.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--vdj_r2_length"
+    description: "Limit the length of the input Read 2 sequence of V(D)J libraries\
+      \ to the first N bases, where N is a user-supplied value. \nTrimming occurs\
+      \ before sequencing metrics are computed and therefore, limiting the length\
+      \ of Read 2 may affect Q30 scores\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--vdj_denovo"
+    description: "Run in reference-free mode (i.e., do not use annotations). This\
+      \ option is not supported for multiplexed experiments.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "3' Cell multiplexing parameters (CellPlex Multiplexing)"
+  arguments:
+  - type: "string"
+    name: "--cell_multiplex_oligo_ids"
+    alternatives:
+    - "--cmo_ids"
+    description: "The Cell Multiplexing oligo IDs used to multiplex this sample. If\
+      \ multiple CMOs were used for a sample,\nseparate IDs with a pipe (e.g., CMO301|CMO302).\
+      \ Required for Cell Multiplexing libraries.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "double"
+    name: "--min_assignment_confidence"
+    description: "The minimum estimated likelihood to call a sample as tagged with\
+      \ a Cell Multiplexing Oligo (CMO) instead of \"Unassigned\".\nUsers may wish\
+      \ to tolerate a higher rate of mis-assignment in order to obtain more singlets\
+      \ to include in their analysis,\nor a lower rate of mis-assignment at the cost\
+      \ of obtaining fewer singlets.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--cmo_set"
+    description: "Path to a custom CMO set CSV file, declaring CMO constructs and\
+      \ associated barcodes. If the default CMO reference IDs that are built into\n\
+      the Cell Ranger software are required, this option does not need to be used.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--barcode_sample_assignment"
+    description: "Path to a barcode-sample assignment CSV file that specifies the\
+      \ barcodes that belong to each sample.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Hashtag multiplexing parameters"
+  arguments:
+  - type: "string"
+    name: "--hashtag_ids"
+    description: "The hashtag IDs used to multiplex this sample. If multiple antibody\
+      \ hashtags were used for the same sample,\nyou can separate IDs with a pipe.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "On-chip multiplexing parameters"
+  arguments:
+  - type: "string"
+    name: "--ocm_barcode_ids"
+    description: "The OCM barcode IDs used to multiplex this sample. Must be one of\
+      \ OB1, OB2, OB3, OB4.\nIf multiple OCM Barcodes were used for the same sample,\
+      \ you can separate IDs\nwith a pipe (e.g., OB1|OB2).\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Flex multiplexing paramaters"
+  arguments:
+  - type: "file"
+    name: "--probe_set"
+    description: "A probe set reference CSV file. It specifies the sequences used\
+      \ as a reference for probe alignment and the gene ID associated with each probe.\n\
+      It must include 4 columns (probe file format 1.0.0): gene_id,probe_seq,probe_id,included,region\
+      \ and an optional 5th column (probe file format 1.0.1).\n- gene_id: The Ensembl\
+      \ gene identifier targeted by the probe.\n- probe_seq: The nucleotide sequence\
+      \ of the probe, which is complementary to the transcript sequence.\n- probe_id:\
+      \ The probe identifier, whose format is described in Probe identifiers.\n- included:\
+      \ A TRUE or FALSE flag specifying whether the probe is included in the filtered\
+      \ counts matrix output or excluded by the probe filter. \n            See filter-probes\
+      \ option of cellranger multi. All probes of a gene must be marked TRUE in the\
+      \ included column for that gene to be included.\n- region: Present only in v1.0.1\
+      \ probe set reference CSV. The gene boundary targeted by the probe. Accepted\
+      \ values are spliced or unspliced.\n\nThe file also contains a number of required\
+      \ metadata fields in the header in the format #key=value:\n- panel_name: The\
+      \ name of the probe set.\n- panel_type: Always predesigned for predesigned probe\
+      \ sets.\n- reference_genome: The reference genome build used for probe design.\n\
+      - reference_version: The version of the Cell Ranger reference transcriptome\
+      \ used for probe design.\n- probe_set_file_format: The version of the probe\
+      \ set file format specification that this file conforms to.\n"
+    info: null
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--filter_probes"
+    description: "If 'false', include all non-deprecated probes listed in the probe\
+      \ set reference CSV file.\nIf 'true' or not set, probes that are predicted to\
+      \ have off-target activity to homologous genes are excluded from analysis.\n\
+      Not filtering will result in UMI counts from all non-deprecated probes,\nincluding\
+      \ those with predicted off-target activity, to be used in the analysis.\nProbes\
+      \ whose ID is prefixed with DEPRECATED are always excluded from the analysis.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--probe_barcode_ids"
+    description: "The Fixed RNA Probe Barcode ID used for this sample, and for multiplex\
+      \ GEX + Antibody Capture libraries,\nthe corresponding Antibody Multiplexing\
+      \ Barcode IDs. 10x recommends specifying both barcodes (e.g., BC001+AB001)\n\
+      when an Antibody Capture library is present. The barcode pair order is BC+AB\
+      \ and they\nare separated with a \"+\" (no spaces). Alternatively, you can specify\
+      \ the Probe Barcode ID alone and\nCell Ranger's barcode pairing auto-detection\
+      \ algorithm will automatically match to the corresponding Antibody\nMultiplexing\
+      \ Barcode.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--emptydrops_minimum_umis"
+    description: "For singleplex Flex experiments, use this option to adjust the UMI\
+      \ cutoff during the second step of cell calling.\nCell Ranger will still perform\
+      \ the full cell calling process but will only evaluate barcodes with UMIs above\n\
+      the threshold you specify.\n"
+    info: null
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Antigen Capture (BEAM) libary arguments"
+  description: "These arguments are recommended if an Antigen Capture (BEAM) library\
+    \ is present. \nIt is needed to calculate the antigen specificity score.\n"
+  arguments:
+  - type: "string"
+    name: "--control_id"
+    description: "A user-defined ID for any negative controls used in the T/BCR Antigen\
+      \ Capture assay. Must match id specified in the feature reference CSV.\nMay\
+      \ only include ASCII characters and must not use whitespace, slash, quote, or\
+      \ comma characters. \nEach ID must be unique and must not collide with a gene\
+      \ identifier from the transcriptome.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--mhc_allele"
+    description: "The MHC allele for TCR Antigen Capture libraries. Must match mhc_allele\
+      \ name specified in the Feature Reference CSV.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "General arguments"
+  description: "These arguments are applicable to all library types.\n"
+  arguments:
+  - type: "boolean"
+    name: "--check_library_compatibility"
+    description: "Optional. This option allows users to disable the check that evaluates\
+      \ 10x Barcode overlap between\nibraries when multiple libraries are specified\
+      \ (e.g., Gene Expression + Antibody Capture). Setting\nthis option to false\
+      \ will disable the check across all library combinations. We recommend running\n\
+      this check (default), however if the pipeline errors out, users can bypass the\
+      \ check to generate\noutputs for troubleshooting.\n"
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  arguments:
+  - type: "file"
+    name: "--output_raw"
+    description: "The raw output folder."
+    info: null
+    example:
+    - "output_dir"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_h5mu"
+    description: "Locations for the output files. Must contain a wildcard (*) character,\n\
+      which will be replaced with the sample name.\n"
+    info: null
+    example:
+    - "*.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--uns_metrics"
+    description: "Name of the .uns slot under which to QC metrics (if any)."
+    info: null
+    default:
+    - "metrics_cellranger"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_ingestion_qc_report"
+    description: "Ingestion QC report in HTML format. Generated when --create_sample_qc_report\
+      \ is true.\n"
+    info: null
+    example:
+    - "*.sample_qc_report.html"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: true
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_processed_h5mu"
+    description: "Folder containing the QC-processed h5mu files. Generated when\n\
+      --create_sample_qc_report is true.\n"
+    info: null
+    example:
+    - "processed_h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--output_multiqc_report"
+    description: "MultiQC report in HTML format. Generated when\n--create_multiqc_report\
+      \ is true.\n"
+    info: null
+    example:
+    - "multiqc_report.html"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+- name: "QC reports"
+  description: "Options for generating optional QC reports."
+  arguments:
+  - type: "boolean"
+    name: "--create_sample_qc_report"
+    description: "Whether to generate an ingestion QC report.\n"
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--create_multiqc_report"
+    description: "Whether to run FastQC on the input FASTQ files and aggregate results\n\
+      into a MultiQC report.\n"
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--run_cellbender"
+    description: "Whether to run CellBender for ambient RNA removal as part of the\n\
+      sample QC report generation. Only used when --create_sample_qc_report is true.\n"
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "A pipeline for running Cell Ranger multi followed by QC."
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "raw"
+- type: "file"
+  path: "raw"
+- type: "file"
+  path: "reference_gencodev41_chr1"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+dependencies:
+- name: "workflows/ingestion/cellranger_multi"
+  repository:
+    type: "vsh"
+    repo: "openpipeline"
+    tag: "v4.0.4"
+- name: "workflows/generate_qc_report"
+  repository:
+    type: "vsh"
+    repo: "openpipeline_qc"
+    tag: "v0.2.2"
+- name: "fastqc"
+  repository:
+    type: "vsh"
+    repo: "biobox"
+    tag: "v0.4.2"
+- name: "multiqc"
+  repository:
+    type: "vsh"
+    repo: "biobox"
+    tag: "v0.4.2"
+repositories:
+- type: "vsh"
+  name: "openpipeline"
+  repo: "openpipeline"
+  tag: "v4.0.4"
+- type: "vsh"
+  name: "openpipeline_qc"
+  repo: "openpipeline_qc"
+  tag: "v0.2.2"
+- type: "vsh"
+  name: "biobox"
+  repo: "biobox"
+  tag: "v0.4.2"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline_composed"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/single_cell/cellranger_multi_qc/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/single_cell/cellranger_multi_qc"
+  executable: "target/nextflow/single_cell/cellranger_multi_qc/main.nf"
+  viash_version: "0.9.7"
+  git_commit: "50682718d61f38677e46dde857fb948846ec6cf6"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline_composed"
+  dependencies:
+  - "target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi"
+  - "target/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report"
+  - "target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc"
+  - "target/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc"
+package_config:
+  name: "openpipeline_composed"
+  version: "set-public-scope"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-bio/openpipeline_incubator/resources_test"
+      dest: "resources_test"
+  repositories:
+  - type: "vsh"
+    name: "openpipeline"
+    repo: "openpipeline"
+    tag: "v4.0.4"
+  - type: "vsh"
+    name: "openpipeline_qc"
+    repo: "openpipeline_qc"
+    tag: "v0.2.2"
+  - type: "vsh"
+    name: "biobox"
+    repo: "biobox"
+    tag: "v0.4.2"
+  viash_version: "0.9.7"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag\
+    \ := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'set-public-scope'"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline_composed"
+    docker_registry: "ghcr.io"
diff --git a/target/nextflow/single_cell/cellranger_multi_qc/main.nf b/target/nextflow/single_cell/cellranger_multi_qc/main.nf
new file mode 100644
index 0000000..1e8785b
--- /dev/null
+++ b/target/nextflow/single_cell/cellranger_multi_qc/main.nf
@@ -0,0 +1,4263 @@
+// cellranger_multi_qc set-public-scope
+// 
+// This wrapper script is auto-generated by viash 0.9.7 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Jakub Majercik (author, maintainer)
+//  * Weiwei Schultz (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  java.nio.file.Files.newBufferedReader(inputFile).withCloseable { br ->
+    def row = 0
+    def header = null
+    def line
+
+    while (header == null && (line = br.readLine()) != null) {
+      if (!line.startsWith("#")) {
+        header = splitRegex.split(line, -1).collect { field ->
+          def m = removeQuote.matcher(field)
+          m.find() ? m.replaceFirst('$1') : field
+        }
+      }
+      row++
+    }
+    assert header != null : "CSV file should contain a header"
+
+    while ((line = br.readLine()) != null) {
+      row++
+      if (!line.startsWith("#")) {
+        def predata = splitRegex.split(line, -1)
+        def data = predata.collect { field ->
+          if (field == "") {
+            return null
+          }
+          def m = removeQuote.matcher(field)
+          if (m.find()) {
+            return m.replaceFirst('$1')
+          } else {
+            return field
+          }
+        }
+        assert header.size() == data.size() : "Row $row should contain the same number as fields as the header"
+
+        def dataMap = [header, data].transpose().collectEntries().findAll { it.value != null }
+        output.add(dataMap)
+      }
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      def infileString = infile.toString()
+      def outfileString = outfile.toString()
+      if (infileString != outfileString) {
+        /* Trailing slashes are removed from both the source and destination arguments.
+           From source arguments, this is useful when a source argument may have a trailing slash
+           and specify a symbolic link to a directory. Without removing the slash, cp will dereference
+           the symbolic link. 
+           See https://www.gnu.org/software/coreutils/manual/html_node/Trailing-slashes.html#Trailing-slashes-1
+
+           For the destination path addding a trailing slash is a problem when publishing directories: 
+           it requires the destination directory to exist. This fails because we only create the parent
+           directories first.
+        */
+        def regexTrailingSlashes = ~/\/+$/
+        def infileNoTrailingSlash = infileString - regexTrailingSlashes
+        def outfileNoTrailingSlash = outfileString - regexTrailingSlashes
+        [
+          "[ -d \"\$(dirname '${outfileNoTrailingSlash}')\" ] || mkdir -p \"\$(dirname '${outfileNoTrailingSlash}')\"",
+          "cp -a '${infileNoTrailingSlash}' '${outfileNoTrailingSlash}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "cellranger_multi_qc",
+  "namespace" : "single_cell",
+  "version" : "set-public-scope",
+  "authors" : [
+    {
+      "name" : "Jakub Majercik",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "links" : {
+          "email" : "jakub@data-intuitive.com",
+          "github" : "jakubmajercik",
+          "linkedin" : "jakubmajercik"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Bioinformatics Engineer"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Input files",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "The FASTQ files to be analyzed. FASTQ files should conform to the naming conventions of bcl2fastq and mkfastq:\n`[Sample Name]_S[Sample Index]_L00[Lane Number]_[Read Type]_001.fastq.gz`\n",
+          "example" : [
+            "mysample_S1_L001_R1_001.fastq.gz",
+            "mysample_S1_L001_R2_001.fastq.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Library arguments",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--library_id",
+          "description" : "The Illumina sample name to analyze. This must exactly match the 'Sample Name'part\nof the FASTQ files specified in the `--input` argument.\n",
+          "example" : [
+            "mysample1"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--library_type",
+          "description" : "The underlying feature type of the library.\n",
+          "example" : [
+            "Gene Expression"
+          ],
+          "required" : false,
+          "choices" : [
+            "Gene Expression",
+            "VDJ",
+            "VDJ-T",
+            "VDJ-B",
+            "VDJ-T-GD",
+            "Antibody Capture",
+            "CRISPR Guide Capture",
+            "Multiplexing Capture",
+            "Antigen Capture",
+            "Custom"
+          ],
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--library_subsample",
+          "description" : "The rate at which reads from the provided FASTQ files are sampled.\nMust be strictly greater than 0 and less than or equal to 1.\n",
+          "example" : [
+            "0.5"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--library_lanes",
+          "description" : "Lanes associated with this sample. Defaults to using all lanes.",
+          "example" : [
+            "1-4"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--library_chemistry",
+          "description" : "Only applicable to FRP. Library-specific assay configuration. By default,\nthe assay configuration is detected automatically. Typically, users will\nnot need to specify a chemistry.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Sample parameters",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--sample_ids",
+          "alternatives" : [
+            "--cell_multiplex_sample_id"
+          ],
+          "description" : "A name to identify a multiplexed sample. Must be alphanumeric with hyphens and/or underscores,\nand less than 64 characters. Required for Cell Multiplexing libraries.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--sample_description",
+          "alternatives" : [
+            "--cell_multiplex_description"
+          ],
+          "description" : "A description for the sample.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--sample_expect_cells",
+          "description" : "Expected number of recovered cells, used as input to cell calling algorithm.\n",
+          "example" : [
+            3000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--sample_force_cells",
+          "description" : "Force pipeline to use this number of cells, bypassing cell detection.\n",
+          "example" : [
+            3000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Feature Barcode library specific arguments",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--feature_reference",
+          "description" : "Path to the Feature reference CSV file, declaring Feature Barcode constructs and associated barcodes.\nRequired only for Antibody Capture or CRISPR Guide Capture libraries.\nSee https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/using/feature-bc-analysis#feature-ref for more information.\\"\n",
+          "example" : [
+            "feature_reference.csv"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--feature_r1_length",
+          "description" : "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,\nwhere N is the user-supplied value. Note that the length includes the Barcode and UMI\nsequences so do not set this below 26.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--feature_r2_length",
+          "description" : "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,\nwhere N is a user-supplied value. Trimming occurs before sequencing metrics are computed\nand therefore, limiting the length of Read 2 may affect Q30 scores.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--min_crispr_umi",
+          "description" : "Set the minimum number of CRISPR guide RNA UMIs required for protospacer detection.\nIf a lower or higher sensitivity is desired for detection, this value can be customized\naccording to specific experimental needs. Applicable only to datasets that include a\nCRISPR Guide Capture library.\n",
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Gene expression arguments",
+      "description" : "Arguments relevant to the analysis of gene expression data.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--gex_reference",
+          "description" : "Genome refence index built by Cell Ranger mkref.",
+          "example" : [
+            "reference_genome.tar.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--gex_secondary_analysis",
+          "description" : "Whether or not to run the secondary analysis e.g. clustering.",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--gex_generate_bam",
+          "description" : "Whether to generate a BAM file.",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--tenx_cloud_token_path",
+          "description" : "The 10x Cloud Analysis user token used to enable cell annotation.",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--cell_annotation_model",
+          "description" : "\\"Cell annotation model to use. If auto, uses the default model for the species.\nIf not given, does not run cell annotation.\\"\n",
+          "required" : false,
+          "choices" : [
+            "auto",
+            "human_pca_v1_beta",
+            "mouse_pca_v1_beta"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gex_expect_cells",
+          "description" : "Expected number of recovered cells, used as input to cell calling algorithm.\n",
+          "example" : [
+            3000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gex_force_cells",
+          "description" : "Force pipeline to use this number of cells, bypassing cell detection.\n",
+          "example" : [
+            3000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--gex_include_introns",
+          "description" : "Whether or not to include intronic reads in counts.\nThis option does not apply to Fixed RNA Profiling analysis.\n",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gex_r1_length",
+          "description" : "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,\nwhere N is the user-supplied value. Note that the length includes the Barcode and UMI\nsequences so do not set this below 26.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--gex_r2_length",
+          "description" : "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,\nwhere N is a user-supplied value. Trimming occurs before sequencing metrics are computed\nand therefore, limiting the length of Read 2 may affect Q30 scores.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--gex_chemistry",
+          "description" : "Assay configuration. Either specify a single value which will be applied to all libraries,\nor a number of values that is equal to the number of libararies. The latter is only applicable\nto only applicable to Fixed RNA Profiling.\n  - auto: Chemistry autodetection (default)\n  - threeprime: Single Cell 3'\n  - SC3Pv1, SC3Pv2, SC3Pv3(-polyA), SC3Pv4(-polyA): Single Cell 3' v1, v2, v3, or v4\n  - SC3Pv3HT(-polyA): Single Cell 3' v3.1 HT\n  - SC-FB: Single Cell Antibody-only 3' v2 or 5'\n  - fiveprime: Single Cell 5'\n  - SC5P-PE: Paired-end Single Cell 5'\n  - SC5P-PE-v3: Paired-end Single Cell 5' v3\n  - SC5P-R2: R2-only Single Cell 5'\n  - SC5P-R2-v3: R2-only Single Cell 5' v3\n  - SCP5-PE-v3: Single Cell 5' paired-end v3 (GEM-X)\n  - SC5PHT : Single Cell 5' v2 HT\n  - SFRP: Fixed RNA Profiling (Singleplex)\n  - MFRP: Fixed RNA Profiling (Multiplex, Probe Barcode on R2)\n  - MFRP-R1: Fixed RNA Profiling (Multiplex, Probe Barcode on R1)\n  - MFRP-RNA: Fixed RNA Profiling (Multiplex, RNA, Probe Barcode on R2)\n  - MFRP-Ab: Fixed RNA Profiling (Multiplex, Antibody, Probe Barcode at R2:69)\n  - MFRP-Ab-R2pos50: Fixed RNA Profiling (Multiplex, Antibody, Probe Barcode at R2:50)\n  - MFRP-RNA-R1: Fixed RNA Profiling (Multiplex, RNA, Probe Barcode on R1)\n  - MFRP-Ab-R1: Fixed RNA Profiling (Multiplex, Antibody, Probe Barcode on R1)\n  - ARC-v1 for analyzing the Gene Expression portion of Multiome data. If Cell Ranger auto-detects ARC-v1 chemistry, an error is triggered.\nSee https://kb.10xgenomics.com/hc/en-us/articles/115003764132-How-does-Cell-Ranger-auto-detect-chemistry- for more information.\n",
+          "default" : [
+            "auto"
+          ],
+          "required" : false,
+          "choices" : [
+            "auto",
+            "threeprime",
+            "fiveprime",
+            "SC3Pv1",
+            "SC3Pv2",
+            "SC3Pv3",
+            "SC3Pv3-polyA",
+            "SC3Pv4",
+            "SC3Pv4-polyA",
+            "SC3Pv3LT",
+            "SC3Pv3HT",
+            "SC3Pv3HT-polyA",
+            "SC5P-PE",
+            "SC5P-PE-v3",
+            "SC5P-R2",
+            "SC-FB",
+            "SC5P-R2-v3",
+            "SCP5-PE-v3",
+            "SC5PHT",
+            "MFRP",
+            "MFRP-R1",
+            "MFRP-RNA",
+            "MFRP-Ab",
+            "SFRP",
+            "MFRP-Ab-R2pos50",
+            "MFRP-RNA-R1",
+            "MFRP-Ab-R1",
+            "ARC-v1"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "VDJ related parameters",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--vdj_reference",
+          "description" : "VDJ refence index built by Cell Ranger mkref.",
+          "example" : [
+            "reference_vdj.tar.gz"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--vdj_inner_enrichment_primers",
+          "description" : "V(D)J Immune Profiling libraries: if inner enrichment primers other than those provided \nin the 10x Genomics kits are used, they need to be specified here as a\ntext file with one primer per line.\n",
+          "example" : [
+            "enrichment_primers.txt"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--vdj_r1_length",
+          "description" : "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases, where N is the user-supplied value.\nNote that the length includes the Barcode and UMI sequences so do not set this below 26.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--vdj_r2_length",
+          "description" : "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases, where N is a user-supplied value. \nTrimming occurs before sequencing metrics are computed and therefore, limiting the length of Read 2 may affect Q30 scores\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--vdj_denovo",
+          "description" : "Run in reference-free mode (i.e., do not use annotations). This option is not supported for multiplexed experiments.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "3' Cell multiplexing parameters (CellPlex Multiplexing)",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--cell_multiplex_oligo_ids",
+          "alternatives" : [
+            "--cmo_ids"
+          ],
+          "description" : "The Cell Multiplexing oligo IDs used to multiplex this sample. If multiple CMOs were used for a sample,\nseparate IDs with a pipe (e.g., CMO301|CMO302). Required for Cell Multiplexing libraries.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--min_assignment_confidence",
+          "description" : "The minimum estimated likelihood to call a sample as tagged with a Cell Multiplexing Oligo (CMO) instead of \\"Unassigned\\".\nUsers may wish to tolerate a higher rate of mis-assignment in order to obtain more singlets to include in their analysis,\nor a lower rate of mis-assignment at the cost of obtaining fewer singlets.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--cmo_set",
+          "description" : "Path to a custom CMO set CSV file, declaring CMO constructs and associated barcodes. If the default CMO reference IDs that are built into\nthe Cell Ranger software are required, this option does not need to be used.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--barcode_sample_assignment",
+          "description" : "Path to a barcode-sample assignment CSV file that specifies the barcodes that belong to each sample.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Hashtag multiplexing parameters",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--hashtag_ids",
+          "description" : "The hashtag IDs used to multiplex this sample. If multiple antibody hashtags were used for the same sample,\nyou can separate IDs with a pipe.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "On-chip multiplexing parameters",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--ocm_barcode_ids",
+          "description" : "The OCM barcode IDs used to multiplex this sample. Must be one of OB1, OB2, OB3, OB4.\nIf multiple OCM Barcodes were used for the same sample, you can separate IDs\nwith a pipe (e.g., OB1|OB2).\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Flex multiplexing paramaters",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--probe_set",
+          "description" : "A probe set reference CSV file. It specifies the sequences used as a reference for probe alignment and the gene ID associated with each probe.\nIt must include 4 columns (probe file format 1.0.0): gene_id,probe_seq,probe_id,included,region and an optional 5th column (probe file format 1.0.1).\n- gene_id: The Ensembl gene identifier targeted by the probe.\n- probe_seq: The nucleotide sequence of the probe, which is complementary to the transcript sequence.\n- probe_id: The probe identifier, whose format is described in Probe identifiers.\n- included: A TRUE or FALSE flag specifying whether the probe is included in the filtered counts matrix output or excluded by the probe filter. \n            See filter-probes option of cellranger multi. All probes of a gene must be marked TRUE in the included column for that gene to be included.\n- region: Present only in v1.0.1 probe set reference CSV. The gene boundary targeted by the probe. Accepted values are spliced or unspliced.\n\nThe file also contains a number of required metadata fields in the header in the format #key=value:\n- panel_name: The name of the probe set.\n- panel_type: Always predesigned for predesigned probe sets.\n- reference_genome: The reference genome build used for probe design.\n- reference_version: The version of the Cell Ranger reference transcriptome used for probe design.\n- probe_set_file_format: The version of the probe set file format specification that this file conforms to.\n",
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--filter_probes",
+          "description" : "If 'false', include all non-deprecated probes listed in the probe set reference CSV file.\nIf 'true' or not set, probes that are predicted to have off-target activity to homologous genes are excluded from analysis.\nNot filtering will result in UMI counts from all non-deprecated probes,\nincluding those with predicted off-target activity, to be used in the analysis.\nProbes whose ID is prefixed with DEPRECATED are always excluded from the analysis.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--probe_barcode_ids",
+          "description" : "The Fixed RNA Probe Barcode ID used for this sample, and for multiplex GEX + Antibody Capture libraries,\nthe corresponding Antibody Multiplexing Barcode IDs. 10x recommends specifying both barcodes (e.g., BC001+AB001)\nwhen an Antibody Capture library is present. The barcode pair order is BC+AB and they\nare separated with a \\"+\\" (no spaces). Alternatively, you can specify the Probe Barcode ID alone and\nCell Ranger's barcode pairing auto-detection algorithm will automatically match to the corresponding Antibody\nMultiplexing Barcode.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--emptydrops_minimum_umis",
+          "description" : "For singleplex Flex experiments, use this option to adjust the UMI cutoff during the second step of cell calling.\nCell Ranger will still perform the full cell calling process but will only evaluate barcodes with UMIs above\nthe threshold you specify.\n",
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Antigen Capture (BEAM) libary arguments",
+      "description" : "These arguments are recommended if an Antigen Capture (BEAM) library is present. \nIt is needed to calculate the antigen specificity score.\n",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--control_id",
+          "description" : "A user-defined ID for any negative controls used in the T/BCR Antigen Capture assay. Must match id specified in the feature reference CSV.\nMay only include ASCII characters and must not use whitespace, slash, quote, or comma characters. \nEach ID must be unique and must not collide with a gene identifier from the transcriptome.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--mhc_allele",
+          "description" : "The MHC allele for TCR Antigen Capture libraries. Must match mhc_allele name specified in the Feature Reference CSV.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "General arguments",
+      "description" : "These arguments are applicable to all library types.\n",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--check_library_compatibility",
+          "description" : "Optional. This option allows users to disable the check that evaluates 10x Barcode overlap between\nibraries when multiple libraries are specified (e.g., Gene Expression + Antibody Capture). Setting\nthis option to false will disable the check across all library combinations. We recommend running\nthis check (default), however if the pipeline errors out, users can bypass the check to generate\noutputs for troubleshooting.\n",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output_raw",
+          "description" : "The raw output folder.",
+          "example" : [
+            "output_dir"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_h5mu",
+          "description" : "Locations for the output files. Must contain a wildcard (*) character,\nwhich will be replaced with the sample name.\n",
+          "example" : [
+            "*.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--uns_metrics",
+          "description" : "Name of the .uns slot under which to QC metrics (if any).",
+          "default" : [
+            "metrics_cellranger"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_ingestion_qc_report",
+          "description" : "Ingestion QC report in HTML format. Generated when --create_sample_qc_report is true.\n",
+          "example" : [
+            "*.sample_qc_report.html"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_processed_h5mu",
+          "description" : "Folder containing the QC-processed h5mu files. Generated when\n--create_sample_qc_report is true.\n",
+          "example" : [
+            "processed_h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--output_multiqc_report",
+          "description" : "MultiQC report in HTML format. Generated when\n--create_multiqc_report is true.\n",
+          "example" : [
+            "multiqc_report.html"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "QC reports",
+      "description" : "Options for generating optional QC reports.",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--create_sample_qc_report",
+          "description" : "Whether to generate an ingestion QC report.\n",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--create_multiqc_report",
+          "description" : "Whether to run FastQC on the input FASTQ files and aggregate results\ninto a MultiQC report.\n",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--run_cellbender",
+          "description" : "Whether to run CellBender for ambient RNA removal as part of the\nsample QC report generation. Only used when --create_sample_qc_report is true.\n",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/configs/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "A pipeline for running Cell Ranger multi followed by QC.",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_anticmv/raw/"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/10x_5k_fixed/raw/"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/reference_gencodev41_chr1"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "dependencies" : [
+    {
+      "name" : "workflows/ingestion/cellranger_multi",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.4"
+      }
+    },
+    {
+      "name" : "workflows/generate_qc_report",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "openpipeline_qc",
+        "tag" : "v0.2.2"
+      }
+    },
+    {
+      "name" : "fastqc",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "biobox",
+        "tag" : "v0.4.2"
+      }
+    },
+    {
+      "name" : "multiqc",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "biobox",
+        "tag" : "v0.4.2"
+      }
+    }
+  ],
+  "repositories" : [
+    {
+      "type" : "vsh",
+      "name" : "openpipeline",
+      "repo" : "openpipeline",
+      "tag" : "v4.0.4"
+    },
+    {
+      "type" : "vsh",
+      "name" : "openpipeline_qc",
+      "repo" : "openpipeline_qc",
+      "tag" : "v0.2.2"
+    },
+    {
+      "type" : "vsh",
+      "name" : "biobox",
+      "repo" : "biobox",
+      "tag" : "v0.4.2"
+    }
+  ],
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline_composed",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/single_cell/cellranger_multi_qc/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/single_cell/cellranger_multi_qc",
+    "viash_version" : "0.9.7",
+    "git_commit" : "50682718d61f38677e46dde857fb948846ec6cf6",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline_composed"
+  },
+  "package_config" : {
+    "name" : "openpipeline_composed",
+    "version" : "set-public-scope",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-bio/openpipeline_incubator/resources_test",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "repositories" : [
+      {
+        "type" : "vsh",
+        "name" : "openpipeline",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.4"
+      },
+      {
+        "type" : "vsh",
+        "name" : "openpipeline_qc",
+        "repo" : "openpipeline_qc",
+        "tag" : "v0.2.2"
+      },
+      {
+        "type" : "vsh",
+        "name" : "biobox",
+        "repo" : "biobox",
+        "tag" : "v0.4.2"
+      }
+    ],
+    "viash_version" : "0.9.7",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'\n",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'set-public-scope'"
+    ],
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline_composed",
+      "docker_registry" : "ghcr.io"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { cellranger_multi } from "${meta.root_dir}/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/ingestion/cellranger_multi/main.nf"
+include { generate_qc_report } from "${meta.root_dir}/dependencies/vsh/vsh/openpipeline_qc/v0.2.2/nextflow/workflows/generate_qc_report/main.nf"
+include { fastqc } from "${meta.root_dir}/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/fastqc/main.nf"
+include { multiqc } from "${meta.root_dir}/dependencies/vsh/vsh/biobox/v0.4.2/nextflow/multiqc/main.nf"
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+    input_ch
+
+  main:
+    output_ch = input_ch
+      | map { id, state ->
+          [id, state + [_meta: [join_id: id]]]
+        }
+
+      | fastqc.run(
+          runIf: { id, state -> state.create_multiqc_report && state.library_type?.contains("Gene Expression") },
+          fromState: { id, state ->
+            [
+              input: state.input,
+              outdir: "${id}_fastqc"
+            ]
+          },
+          toState: { id, output, state ->
+            state + [output_fastqc: output.outdir]
+          }
+        )
+
+      | cellranger_multi.run(
+          fromState: { id, state -> state },
+          toState: { id, output, state ->
+            state + [
+              output_raw: output.output_raw,
+              output_h5mu: output.output_h5mu
+            ]
+          }
+        )
+
+      | multiqc.run(
+          runIf: { id, state -> state.create_multiqc_report && state.library_type?.contains("Gene Expression") },
+          fromState: { id, state ->
+            [
+              input: [state.output_fastqc, state.output_raw],
+              output_report: state.output_multiqc_report
+            ]
+          },
+          toState: { id, output, state ->
+            state + [_multiqc_produced: true, output_multiqc_report: output.output_report]
+          }
+        )
+      
+      | generate_qc_report.run(
+          runIf: { id, state -> state.create_sample_qc_report && state.library_type?.contains("Gene Expression") },
+          fromState: { id, state ->
+            [
+              id: id,
+              input: state.output_h5mu,
+              ingestion_method: "cellranger_multi",
+              run_cellbender: state.run_cellbender,
+              output_qc_report: state.output_ingestion_qc_report,
+              output_processed_h5mu: state.output_processed_h5mu
+            ]
+          },
+          toState: { id, output, state ->
+            state + [
+              _qc_report_produced: true,
+              output_ingestion_qc_report: output.output_qc_report,
+              output_processed_h5mu: output.output_processed_h5mu
+            ]
+          }
+        )
+
+      | map { id, state ->
+          def out = [output_raw: state.output_raw, output_h5mu: state.output_h5mu]
+          if (state._meta) out._meta = state._meta
+          if (state._multiqc_produced) out.output_multiqc_report = state.output_multiqc_report
+          if (state._qc_report_produced) {
+            out.output_ingestion_qc_report = state.output_ingestion_qc_report
+            out.output_processed_h5mu = state.output_processed_h5mu
+          }
+          [id, out]
+        }
+
+  emit:
+    output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/nextflow/single_cell/cellranger_multi_qc/nextflow.config b/target/nextflow/single_cell/cellranger_multi_qc/nextflow.config
new file mode 100644
index 0000000..2c12d51
--- /dev/null
+++ b/target/nextflow/single_cell/cellranger_multi_qc/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'single_cell/cellranger_multi_qc'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'set-public-scope'
+  description = 'A pipeline for running Cell Ranger multi followed by QC.'
+  author = 'Jakub Majercik, Weiwei Schultz'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/nextflow/single_cell/cellranger_multi_qc/nextflow_labels.config b/target/nextflow/single_cell/cellranger_multi_qc/nextflow_labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/target/nextflow/single_cell/cellranger_multi_qc/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/nextflow/single_cell/cellranger_multi_qc/nextflow_schema.json b/target/nextflow/single_cell/cellranger_multi_qc/nextflow_schema.json
new file mode 100644
index 0000000..dd50d5e
--- /dev/null
+++ b/target/nextflow/single_cell/cellranger_multi_qc/nextflow_schema.json
@@ -0,0 +1,537 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "cellranger_multi_qc",
+  "description": "A pipeline for running Cell Ranger multi followed by QC.",
+  "type": "object",
+  "$defs": {
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "output_raw": {
+          "type": "string",
+          "format": "path",
+          "description": "The raw output folder.",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output_raw\"`, direction: `output`, example: `\"output_dir\"`. ",
+          "default": "$id.$key.output_raw"
+        },
+        "output_h5mu": {
+          "type": "string",
+          "format": "path",
+          "description": "Locations for the output files",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output_h5mu.h5mu\"`, direction: `output`, example: `\"*.h5mu\"`. ",
+          "default": "$id.$key.output_h5mu.h5mu"
+        },
+        "uns_metrics": {
+          "type": "string",
+          "description": "Name of the .uns slot under which to QC metrics (if any).",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"metrics_cellranger\"`. ",
+          "default": "metrics_cellranger"
+        },
+        "output_ingestion_qc_report": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "Ingestion QC report in HTML format",
+          "help_text": "Type: `file`, multiple: `True`, default: `\"$id.$key.output_ingestion_qc_report_*.html\"`, direction: `output`, example: `[\"*.sample_qc_report.html\"]`. ",
+          "default": "$id.$key.output_ingestion_qc_report_*.html"
+        },
+        "output_processed_h5mu": {
+          "type": "string",
+          "format": "path",
+          "description": "Folder containing the QC-processed h5mu files",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output_processed_h5mu\"`, direction: `output`, example: `\"processed_h5mu\"`. ",
+          "default": "$id.$key.output_processed_h5mu"
+        },
+        "output_multiqc_report": {
+          "type": "string",
+          "format": "path",
+          "description": "MultiQC report in HTML format",
+          "help_text": "Type: `file`, multiple: `False`, default: `\"$id.$key.output_multiqc_report.html\"`, direction: `output`, example: `\"multiqc_report.html\"`. ",
+          "default": "$id.$key.output_multiqc_report.html"
+        }
+      }
+    },
+    "input files": {
+      "title": "Input files",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "input": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "format": "path",
+          "description": "The FASTQ files to be analyzed",
+          "help_text": "Type: `file`, multiple: `True`, direction: `input`, example: `[\"mysample_S1_L001_R1_001.fastq.gz\";\"mysample_S1_L001_R2_001.fastq.gz\"]`. "
+        }
+      }
+    },
+    "library arguments": {
+      "title": "Library arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "library_id": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The Illumina sample name to analyze",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"mysample1\"]`. "
+        },
+        "library_type": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The underlying feature type of the library.\n",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"Gene Expression\"]`, choices: ``Gene Expression`, `VDJ`, `VDJ-T`, `VDJ-B`, `VDJ-T-GD`, `Antibody Capture`, `CRISPR Guide Capture`, `Multiplexing Capture`, `Antigen Capture`, `Custom``. "
+        },
+        "library_subsample": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The rate at which reads from the provided FASTQ files are sampled.\nMust be strictly greater than 0 and less than or equal to 1.\n",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"0.5\"]`. "
+        },
+        "library_lanes": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Lanes associated with this sample",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"1-4\"]`. "
+        },
+        "library_chemistry": {
+          "type": "string",
+          "description": "Only applicable to FRP",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "sample parameters": {
+      "title": "Sample parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "sample_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "A name to identify a multiplexed sample",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "sample_description": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "A description for the sample.",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "sample_expect_cells": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Expected number of recovered cells, used as input to cell calling algorithm.\n",
+          "help_text": "Type: `integer`, multiple: `True`, example: `[3000]`. "
+        },
+        "sample_force_cells": {
+          "type": "array",
+          "items": {
+            "type": "integer"
+          },
+          "description": "Force pipeline to use this number of cells, bypassing cell detection.\n",
+          "help_text": "Type: `integer`, multiple: `True`, example: `[3000]`. "
+        }
+      }
+    },
+    "feature barcode library specific arguments": {
+      "title": "Feature Barcode library specific arguments",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "feature_reference": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to the Feature reference CSV file, declaring Feature Barcode constructs and associated barcodes.\nRequired only for Antibody Capture or CRISPR Guide Capture libraries.\nSee https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/using/feature-bc-analysis#feature-ref for more information.\"\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"feature_reference.csv\"`. "
+        },
+        "feature_r1_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,\nwhere N is the user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "feature_r2_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,\nwhere N is a user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "min_crispr_umi": {
+          "type": "integer",
+          "description": "Set the minimum number of CRISPR guide RNA UMIs required for protospacer detection.\nIf a lower or higher sensitivity is desired for detection, this value can be customized\naccording to specific experimental needs",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        }
+      }
+    },
+    "gene expression arguments": {
+      "title": "Gene expression arguments",
+      "type": "object",
+      "description": "Arguments relevant to the analysis of gene expression data.",
+      "properties": {
+        "gex_reference": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Genome refence index built by Cell Ranger mkref.",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"reference_genome.tar.gz\"`. "
+        },
+        "gex_secondary_analysis": {
+          "type": "boolean",
+          "description": "Whether or not to run the secondary analysis e.g",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "gex_generate_bam": {
+          "type": "boolean",
+          "description": "Whether to generate a BAM file.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "tenx_cloud_token_path": {
+          "type": "string",
+          "format": "path",
+          "description": "The 10x Cloud Analysis user token used to enable cell annotation.",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`. "
+        },
+        "cell_annotation_model": {
+          "type": "string",
+          "description": "\"Cell annotation model to use",
+          "help_text": "Type: `string`, multiple: `False`, choices: ``auto`, `human_pca_v1_beta`, `mouse_pca_v1_beta``. ",
+          "enum": [
+            "auto",
+            "human_pca_v1_beta",
+            "mouse_pca_v1_beta"
+          ]
+        },
+        "gex_expect_cells": {
+          "type": "integer",
+          "description": "Expected number of recovered cells, used as input to cell calling algorithm.\n",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3000`. "
+        },
+        "gex_force_cells": {
+          "type": "integer",
+          "description": "Force pipeline to use this number of cells, bypassing cell detection.\n",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3000`. "
+        },
+        "gex_include_introns": {
+          "type": "boolean",
+          "description": "Whether or not to include intronic reads in counts.\nThis option does not apply to Fixed RNA Profiling analysis.\n",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "gex_r1_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases,\nwhere N is the user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "gex_r2_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases,\nwhere N is a user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "gex_chemistry": {
+          "type": "string",
+          "description": "Assay configuration",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"auto\"`, choices: ``auto`, `threeprime`, `fiveprime`, `SC3Pv1`, `SC3Pv2`, `SC3Pv3`, `SC3Pv3-polyA`, `SC3Pv4`, `SC3Pv4-polyA`, `SC3Pv3LT`, `SC3Pv3HT`, `SC3Pv3HT-polyA`, `SC5P-PE`, `SC5P-PE-v3`, `SC5P-R2`, `SC-FB`, `SC5P-R2-v3`, `SCP5-PE-v3`, `SC5PHT`, `MFRP`, `MFRP-R1`, `MFRP-RNA`, `MFRP-Ab`, `SFRP`, `MFRP-Ab-R2pos50`, `MFRP-RNA-R1`, `MFRP-Ab-R1`, `ARC-v1``. ",
+          "enum": [
+            "auto",
+            "threeprime",
+            "fiveprime",
+            "SC3Pv1",
+            "SC3Pv2",
+            "SC3Pv3",
+            "SC3Pv3-polyA",
+            "SC3Pv4",
+            "SC3Pv4-polyA",
+            "SC3Pv3LT",
+            "SC3Pv3HT",
+            "SC3Pv3HT-polyA",
+            "SC5P-PE",
+            "SC5P-PE-v3",
+            "SC5P-R2",
+            "SC-FB",
+            "SC5P-R2-v3",
+            "SCP5-PE-v3",
+            "SC5PHT",
+            "MFRP",
+            "MFRP-R1",
+            "MFRP-RNA",
+            "MFRP-Ab",
+            "SFRP",
+            "MFRP-Ab-R2pos50",
+            "MFRP-RNA-R1",
+            "MFRP-Ab-R1",
+            "ARC-v1"
+          ],
+          "default": "auto"
+        }
+      }
+    },
+    "vdj related parameters": {
+      "title": "VDJ related parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "vdj_reference": {
+          "type": "string",
+          "format": "path",
+          "description": "VDJ refence index built by Cell Ranger mkref.",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"reference_vdj.tar.gz\"`. "
+        },
+        "vdj_inner_enrichment_primers": {
+          "type": "string",
+          "format": "path",
+          "description": "V(D)J Immune Profiling libraries: if inner enrichment primers other than those provided \nin the 10x Genomics kits are used, they need to be specified here as a\ntext file with one primer per line.\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"enrichment_primers.txt\"`. "
+        },
+        "vdj_r1_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 1 sequence of V(D)J libraries to the first N bases, where N is the user-supplied value.\nNote that the length includes the Barcode and UMI sequences so do not set this below 26.\n",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "vdj_r2_length": {
+          "type": "integer",
+          "description": "Limit the length of the input Read 2 sequence of V(D)J libraries to the first N bases, where N is a user-supplied value",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "vdj_denovo": {
+          "type": "boolean",
+          "description": "Run in reference-free mode (i.e., do not use annotations)",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        }
+      }
+    },
+    "3' cell multiplexing parameters (cellplex multiplexing)": {
+      "title": "3' Cell multiplexing parameters (CellPlex Multiplexing)",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "cell_multiplex_oligo_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The Cell Multiplexing oligo IDs used to multiplex this sample",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "min_assignment_confidence": {
+          "type": "number",
+          "description": "The minimum estimated likelihood to call a sample as tagged with a Cell Multiplexing Oligo (CMO) instead of \"Unassigned\".\nUsers may wish to tolerate a higher rate of mis-assignment in order to obtain more singlets to include in their analysis,\nor a lower rate of mis-assignment at the cost of obtaining fewer singlets.\n",
+          "help_text": "Type: `double`, multiple: `False`. "
+        },
+        "cmo_set": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to a custom CMO set CSV file, declaring CMO constructs and associated barcodes",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`. "
+        },
+        "barcode_sample_assignment": {
+          "type": "string",
+          "format": "path",
+          "description": "Path to a barcode-sample assignment CSV file that specifies the barcodes that belong to each sample.\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`. "
+        }
+      }
+    },
+    "hashtag multiplexing parameters": {
+      "title": "Hashtag multiplexing parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "hashtag_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The hashtag IDs used to multiplex this sample",
+          "help_text": "Type: `string`, multiple: `True`. "
+        }
+      }
+    },
+    "on-chip multiplexing parameters": {
+      "title": "On-chip multiplexing parameters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "ocm_barcode_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The OCM barcode IDs used to multiplex this sample",
+          "help_text": "Type: `string`, multiple: `True`. "
+        }
+      }
+    },
+    "flex multiplexing paramaters": {
+      "title": "Flex multiplexing paramaters",
+      "type": "object",
+      "description": "No description",
+      "properties": {
+        "probe_set": {
+          "type": "string",
+          "format": "path",
+          "description": "A probe set reference CSV file",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`. "
+        },
+        "filter_probes": {
+          "type": "boolean",
+          "description": "If 'false', include all non-deprecated probes listed in the probe set reference CSV file.\nIf 'true' or not set, probes that are predicted to have off-target activity to homologous genes are excluded from analysis.\nNot filtering will result in UMI counts from all non-deprecated probes,\nincluding those with predicted off-target activity, to be used in the analysis.\nProbes whose ID is prefixed with DEPRECATED are always excluded from the analysis.\n",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        },
+        "probe_barcode_ids": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The Fixed RNA Probe Barcode ID used for this sample, and for multiplex GEX + Antibody Capture libraries,\nthe corresponding Antibody Multiplexing Barcode IDs",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "emptydrops_minimum_umis": {
+          "type": "integer",
+          "description": "For singleplex Flex experiments, use this option to adjust the UMI cutoff during the second step of cell calling.\nCell Ranger will still perform the full cell calling process but will only evaluate barcodes with UMIs above\nthe threshold you specify.\n",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        }
+      }
+    },
+    "antigen capture (beam) libary arguments": {
+      "title": "Antigen Capture (BEAM) libary arguments",
+      "type": "object",
+      "description": "These arguments are recommended if an Antigen Capture (BEAM) library is present. \nIt is needed to calculate the antigen specificity score.\n",
+      "properties": {
+        "control_id": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "A user-defined ID for any negative controls used in the T/BCR Antigen Capture assay",
+          "help_text": "Type: `string`, multiple: `True`. "
+        },
+        "mhc_allele": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The MHC allele for TCR Antigen Capture libraries",
+          "help_text": "Type: `string`, multiple: `True`. "
+        }
+      }
+    },
+    "general arguments": {
+      "title": "General arguments",
+      "type": "object",
+      "description": "These arguments are applicable to all library types.\n",
+      "properties": {
+        "check_library_compatibility": {
+          "type": "boolean",
+          "description": "Optional",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        }
+      }
+    },
+    "qc reports": {
+      "title": "QC reports",
+      "type": "object",
+      "description": "Options for generating optional QC reports.",
+      "properties": {
+        "create_sample_qc_report": {
+          "type": "boolean",
+          "description": "Whether to generate an ingestion QC report.\n",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "create_multiqc_report": {
+          "type": "boolean",
+          "description": "Whether to run FastQC on the input FASTQ files and aggregate results\ninto a MultiQC report.\n",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "run_cellbender": {
+          "type": "boolean",
+          "description": "Whether to run CellBender for ambient RNA removal as part of the\nsample QC report generation",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/input files"
+    },
+    {
+      "$ref": "#/$defs/library arguments"
+    },
+    {
+      "$ref": "#/$defs/sample parameters"
+    },
+    {
+      "$ref": "#/$defs/feature barcode library specific arguments"
+    },
+    {
+      "$ref": "#/$defs/gene expression arguments"
+    },
+    {
+      "$ref": "#/$defs/vdj related parameters"
+    },
+    {
+      "$ref": "#/$defs/3' cell multiplexing parameters (cellplex multiplexing)"
+    },
+    {
+      "$ref": "#/$defs/hashtag multiplexing parameters"
+    },
+    {
+      "$ref": "#/$defs/on-chip multiplexing parameters"
+    },
+    {
+      "$ref": "#/$defs/flex multiplexing paramaters"
+    },
+    {
+      "$ref": "#/$defs/antigen capture (beam) libary arguments"
+    },
+    {
+      "$ref": "#/$defs/general arguments"
+    },
+    {
+      "$ref": "#/$defs/qc reports"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}
diff --git a/target/nextflow/single_cell/process_integrate_annotate/.config.vsh.yaml b/target/nextflow/single_cell/process_integrate_annotate/.config.vsh.yaml
new file mode 100644
index 0000000..88b4d9c
--- /dev/null
+++ b/target/nextflow/single_cell/process_integrate_annotate/.config.vsh.yaml
@@ -0,0 +1,839 @@
+name: "process_integrate_annotate"
+namespace: "single_cell"
+version: "set-public-scope"
+authors:
+- name: "Dorien Roosen"
+  roles:
+  - "author"
+  - "maintainer"
+  info:
+    role: "Core Team Member"
+    links:
+      email: "dorien@data-intuitive.com"
+      github: "dorien-er"
+      linkedin: "dorien-roosen"
+    organizations:
+    - name: "Data Intuitive"
+      href: "https://www.data-intuitive.com"
+      role: "Data Scientist"
+- name: "Weiwei Schultz"
+  roles:
+  - "contributor"
+  info:
+    role: "Contributor"
+    organizations:
+    - name: "Janssen R&D US"
+      role: "Associate Director Data Sciences"
+argument_groups:
+- name: "Input (query) data arguments"
+  description: "The input query dataset(s) to be annotated"
+  arguments:
+  - type: "string"
+    name: "--id"
+    description: "ID of the sample."
+    info: null
+    example:
+    - "foo"
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "file"
+    name: "--input"
+    description: "Input query dataset(s) to be annotated"
+    info: null
+    example:
+    - "input.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--modality"
+    description: "Modality to be processed. Should match the modality in the --reference\
+      \ dataset, if provided."
+    info: null
+    default:
+    - "rna"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_layer"
+    description: "The layer in the input data containing the raw counts, if .X is\
+      \ not to be used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--input_var_gene_names"
+    description: "The name of the adata var column containing gene names; when no\
+      \ gene_name_layer is provided, the var index will be used.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--input_reference_gene_overlap"
+    description: "The minimum number of genes present in both the reference and query\
+      \ datasets.\n"
+    info: null
+    default:
+    - 100
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Reference data arguments"
+  description: "Dataset to be used as a reference for label transfer and to train\
+    \ annotation algorithms on."
+  arguments:
+  - type: "file"
+    name: "--reference"
+    description: "The reference dataset in .h5mu format to be used as a reference\
+      \ mapper and to train annotation algorithms on.\n"
+    info: null
+    example:
+    - "reference.h5mu"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_layer_raw_counts"
+    description: "The layer in the reference dataset containing the raw counts, if\
+      \ .X is not to be used."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_layer_lognormalized_counts"
+    description: "The layer in the reference dataset containing the log-normalized\
+      \ counts, if .X is not to be used."
+    info: null
+    default:
+    - "log_normalized"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_var_gene_names"
+    description: "The name of the adata .var column containing gene names if the .var\
+      \ index is not to be used.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_obs_batch"
+    description: "The .obs column of the reference dataset containing the batch information.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_obs_label"
+    description: "The `.obs` key of the target labels to tranfer."
+    info: null
+    example:
+    - "cell_type"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_obs_label_unlabeled_category"
+    description: "Value in the --reference_obs_label field that indicates unlabeled\
+      \ observations"
+    info: null
+    default:
+    - "Unkown"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--reference_var_input"
+    description: ".var column containing highly variable genes. By default, do not\
+      \ subset genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Methods"
+  description: "The available annotation and integration methods to integrate and/or\
+    \ annotate the query dataset(s) with."
+  arguments:
+  - type: "string"
+    name: "--integration_methods"
+    description: "Integration methods to be executed."
+    info: null
+    example:
+    - "harmony;scvi"
+    required: false
+    choices:
+    - "harmony"
+    - "scvi"
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--annotation_methods"
+    description: "Annotation methods to be executed."
+    info: null
+    example:
+    - "celltypist;scanvi_scarches"
+    required: false
+    choices:
+    - "celltypist"
+    - "scanvi_scarches"
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Pre-processing options: RNA filtering"
+  description: "Pre-processing options for filtering RNA data"
+  arguments:
+  - type: "integer"
+    name: "--rna_min_counts"
+    description: "Minimum number of counts captured per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--rna_max_counts"
+    description: "Maximum number of counts captured per cell."
+    info: null
+    example:
+    - 5000000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--rna_min_genes_per_cell"
+    description: "Minimum of non-zero values per cell."
+    info: null
+    example:
+    - 200
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--rna_max_genes_per_cell"
+    description: "Maximum of non-zero values per cell."
+    info: null
+    example:
+    - 1500000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--rna_min_cells_per_gene"
+    description: "Minimum of non-zero values per gene."
+    info: null
+    example:
+    - 3
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--rna_min_fraction_mito"
+    description: "Minimum fraction of UMIs that are mitochondrial."
+    info: null
+    example:
+    - 0.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--rna_max_fraction_mito"
+    description: "Maximum fraction of UMIs that are mitochondrial."
+    info: null
+    example:
+    - 0.2
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Pre-processing options: Highly variable features detection"
+  description: "Pre-processing options for detecting highly variable features"
+  arguments:
+  - type: "integer"
+    name: "--n_hvg"
+    description: "Number of highly-variable features to keep. \nOnly relevant if HVG\
+      \ need to be calculated across query and reference datasets (e.g. for --annotation_methods\
+      \ scvi_knn and harmony_knn). \nFor reference mapping-based methods, the HVG's\
+      \ specified in --reference_var_input will be used.\n"
+    info: null
+    default:
+    - 2000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Pre-processing options: Mitochondrial & Ribosomal Gene Detection"
+  description: "Pre-processing options for detecting mitochondrial genes"
+  arguments:
+  - type: "string"
+    name: "--var_name_mitochondrial_genes"
+    description: "In which .var slot to store a boolean array corresponding the mitochondrial\
+      \ genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--var_name_ribosomal_genes"
+    description: "In which .var slot to store a boolean array corresponding the ribosomal\
+      \ genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_mitochondrial_fraction"
+    description: "When specified, write the fraction of counts originating from mitochondrial\
+      \ genes \n(based on --mitochondrial_gene_regex) to an .obs column with the specified\
+      \ name.\nRequires --var_name_mitochondrial_genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--obs_name_ribosomal_fraction"
+    description: "When specified, write the fraction of counts originating from ribosomal\
+      \ genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified\
+      \ name.\nRequires --var_name_ribosomal_genes.\n"
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--mitochondrial_gene_regex"
+    description: "Regex string that identifies mitochondrial genes from --var_gene_names.\n\
+      By default will detect human and mouse mitochondrial genes from a gene symbol.\n"
+    info: null
+    default:
+    - "^[mM][tT]-"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--ribosomal_gene_regex"
+    description: "Regex string that identifies ribosomal genes from --var_gene_names.\n\
+      By default will detect human and mouse ribosomal genes from a gene symbol.\n"
+    info: null
+    default:
+    - "^[Mm]?[Rr][Pp][LlSs]"
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Pre-processing options: QC metrics calculation options"
+  description: "Pre-processing options for calculating QC metrics"
+  arguments:
+  - type: "string"
+    name: "--var_qc_metrics"
+    description: "Keys to select a boolean (containing only True or False) column\
+      \ from .var.\nFor each cell, calculate the proportion of total values for genes\
+      \ which are labeled 'True', \ncompared to the total sum of the values for all\
+      \ genes. Defaults to the combined values specified for\n--var_name_mitochondrial_genes\
+      \ and --highly_variable_features_var_output.\n"
+    info: null
+    example:
+    - "ercc,highly_variable"
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ","
+- name: "Harmony integration options"
+  description: "Specifications for harmony integration."
+  arguments:
+  - type: "double"
+    name: "--harmony_theta"
+    description: "Diversity clustering penalty parameter. Specify for each variable\
+      \ in group.by.vars. \ntheta=0 does not encourage any diversity. Larger values\
+      \ of theta\nresult in more diverse clusters.\"\n"
+    info: null
+    example:
+    - 0.0
+    - 1.0
+    - 2.0
+    default:
+    - 2.0
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+  - type: "string"
+    name: "--harmony_obs_covariates"
+    description: "The .obs field(s) that define the covariate(s) to regress out."
+    info: null
+    example:
+    - "batch"
+    - "sample"
+    default:
+    - "sample_id"
+    required: true
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "scVI, scANVI and scArches training options"
+  description: "Training arguments for scVI, scANVI and scArches. Relevant for --annotation_methods\
+    \ 'scvi_knn' and 'scanvi_scarches'."
+  arguments:
+  - type: "boolean"
+    name: "--early_stopping"
+    description: "Whether to perform early stopping with respect to the validation\
+      \ set."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "string"
+    name: "--early_stopping_monitor"
+    description: "Metric logged during validation set epoch."
+    info: null
+    default:
+    - "elbo_validation"
+    required: false
+    choices:
+    - "elbo_validation"
+    - "reconstruction_loss_validation"
+    - "kl_local_validation"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--early_stopping_patience"
+    description: "Number of validation epochs with no improvement after which training\
+      \ will be stopped."
+    info: null
+    default:
+    - 45
+    required: false
+    min: 1
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--early_stopping_min_delta"
+    description: "Minimum change in the monitored quantity to qualify as an improvement,\
+      \ i.e. an absolute change of less than min_delta, will count as no improvement."
+    info: null
+    default:
+    - 0.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--max_epochs"
+    description: "Number of passes through the dataset, defaults to (20000 / number\
+      \ of cells) * 400 or 400; whichever is smallest."
+    info: null
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--reduce_lr_on_plateau"
+    description: "Whether to monitor validation loss and reduce learning rate when\
+      \ validation set `lr_scheduler_metric` plateaus."
+    info: null
+    default:
+    - true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--lr_factor"
+    description: "Factor to reduce learning rate."
+    info: null
+    default:
+    - 0.6
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--lr_patience"
+    description: "Number of epochs with no improvement after which learning rate will\
+      \ be reduced."
+    info: null
+    default:
+    - 30.0
+    required: false
+    min: 0.0
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "CellTypist reference model"
+  description: "The CellTypist reference model to use for annotation. If not provided,\
+    \ the reference dataset will be used for model training."
+  arguments:
+  - type: "file"
+    name: "--celltypist_model"
+    description: "Pretrained model in pkl format. If not provided, the model will\
+      \ be trained on the reference data and --reference should be provided."
+    info: null
+    example:
+    - "pretrained_model.pkl"
+    must_exist: true
+    create_parent: true
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "CellTypist annotation options"
+  description: "Specifications for CellTypist annotation."
+  arguments:
+  - type: "boolean"
+    name: "--celltypist_feature_selection"
+    description: "Whether to perform feature selection."
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean"
+    name: "--celltypist_majority_voting"
+    description: "Whether to refine the predicted labels by running the majority voting\
+      \ classifier after over-clustering."
+    info: null
+    default:
+    - false
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "double"
+    name: "--celltypist_C"
+    description: "Inverse of regularization strength in logistic regression."
+    info: null
+    default:
+    - 1.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--celltypist_max_iter"
+    description: "Maximum number of iterations before reaching the minimum of the\
+      \ cost function."
+    info: null
+    default:
+    - 1000
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "boolean_true"
+    name: "--celltypist_use_SGD"
+    description: "Whether to use the stochastic gradient descent algorithm."
+    info: null
+    direction: "input"
+  - type: "double"
+    name: "--celltypist_min_prop"
+    description: "\"For the dominant cell type within a subcluster, the minimum proportion\
+      \ of cells required to \nsupport naming of the subcluster by this cell type.\
+      \ Ignored if majority_voting is set to False. \nSubcluster that fails to pass\
+      \ this proportion threshold will be assigned 'Heterogeneous'.\"\n"
+    info: null
+    default:
+    - 0.0
+    required: false
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Clustering options"
+  description: "Arguments for Leiden clustering. Only relevant for --annotation_methods\
+    \ `scvi_knn`, `scanvi_scarches` and `harmony_knn`."
+  arguments:
+  - type: "double"
+    name: "--leiden_resolution"
+    description: "Control the coarseness of the clustering. Higher values lead to\
+      \ more clusters."
+    info: null
+    default:
+    - 1.0
+    required: false
+    direction: "input"
+    multiple: true
+    multiple_sep: ";"
+- name: "Neighbor classifier arguments"
+  description: "Arguments related to calculating the n nearest neighbors. Only relevant\
+    \ for --annotation_methods `scvi_knn`, `scanvi_scarches` and `harmony_knn`."
+  arguments:
+  - type: "string"
+    name: "--knn_weights"
+    description: "Weight function used in prediction. Possible values are:\n`uniform`\
+      \ (all points in each neighborhood are weighted equally) or \n`distance` (weight\
+      \ points by the inverse of their distance)\n"
+    info: null
+    default:
+    - "uniform"
+    required: false
+    choices:
+    - "uniform"
+    - "distance"
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+  - type: "integer"
+    name: "--knn_n_neighbors"
+    description: "The number of neighbors to use in k-neighbor graph structure used\
+      \ for fast approximate nearest neighbor search with PyNNDescent. \nLarger values\
+      \ will result in more accurate search results at the cost of computation time.\n"
+    info: null
+    default:
+    - 15
+    required: false
+    min: 5
+    direction: "input"
+    multiple: false
+    multiple_sep: ";"
+- name: "Outputs"
+  description: "The output file to write the annotated dataset to."
+  arguments:
+  - type: "file"
+    name: "--output"
+    description: "The output file.\n"
+    info: null
+    example:
+    - "output.h5mu"
+    must_exist: true
+    create_parent: true
+    required: true
+    direction: "output"
+    multiple: false
+    multiple_sep: ";"
+resources:
+- type: "nextflow_script"
+  path: "main.nf"
+  is_executable: true
+  entrypoint: "run_wf"
+- type: "file"
+  path: "nextflow_labels.config"
+  dest: "nextflow_labels.config"
+description: "A pipeline to process, integrate and annotate single cell (multi-)omics\
+  \ data.\nAvailable integration methods:\n- Harmony\n- scVI\nAvailable annotation\
+  \ methods:\n- CellTypist\n- scANVI (with scArches)\n"
+test_resources:
+- type: "nextflow_script"
+  path: "test.nf"
+  is_executable: true
+  entrypoint: "test_wf"
+- type: "file"
+  path: "pbmc_1k_protein_v3_mms.h5mu"
+- type: "file"
+  path: "TS_Blood_filtered.h5mu"
+- type: "file"
+  path: "celltypist_model_Immune_All_Low.pkl"
+info: null
+status: "enabled"
+scope:
+  image: "public"
+  target: "public"
+requirements:
+  commands:
+  - "ps"
+dependencies:
+- name: "workflows/multiomics/process_samples"
+  alias: "process_samples_workflow"
+  repository:
+    type: "vsh"
+    repo: "openpipeline"
+    tag: "v4.0.4"
+- name: "annotate/celltypist"
+  alias: "celltypist_annotation"
+  repository:
+    type: "vsh"
+    repo: "openpipeline"
+    tag: "v4.0.4"
+- name: "workflows/annotation/scanvi_scarches"
+  alias: "scanvi_scarches_annotation"
+  repository:
+    type: "vsh"
+    repo: "openpipeline"
+    tag: "v4.0.4"
+- name: "workflows/integration/harmony_leiden"
+  alias: "harmony_integration"
+  repository:
+    type: "vsh"
+    repo: "openpipeline"
+    tag: "v4.0.4"
+- name: "workflows/integration/scvi_leiden"
+  alias: "scvi_integration"
+  repository:
+    type: "vsh"
+    repo: "openpipeline"
+    tag: "v4.0.4"
+repositories:
+- type: "vsh"
+  name: "openpipeline"
+  repo: "openpipeline"
+  tag: "v4.0.4"
+- type: "vsh"
+  name: "openpipeline_qc"
+  repo: "openpipeline_qc"
+  tag: "v0.2.2"
+- type: "vsh"
+  name: "biobox"
+  repo: "biobox"
+  tag: "v0.4.2"
+links:
+  repository: "https://github.com/openpipelines-bio/openpipeline_composed"
+  docker_registry: "ghcr.io"
+runners:
+- type: "nextflow"
+  id: "nextflow"
+  directives:
+    tag: "$id"
+  auto:
+    simplifyInput: true
+    simplifyOutput: false
+    transcript: false
+    publish: false
+  config:
+    labels:
+      mem1gb: "memory = 1000000000.B"
+      mem2gb: "memory = 2000000000.B"
+      mem5gb: "memory = 5000000000.B"
+      mem10gb: "memory = 10000000000.B"
+      mem20gb: "memory = 20000000000.B"
+      mem50gb: "memory = 50000000000.B"
+      mem100gb: "memory = 100000000000.B"
+      mem200gb: "memory = 200000000000.B"
+      mem500gb: "memory = 500000000000.B"
+      mem1tb: "memory = 1000000000000.B"
+      mem2tb: "memory = 2000000000000.B"
+      mem5tb: "memory = 5000000000000.B"
+      mem10tb: "memory = 10000000000000.B"
+      mem20tb: "memory = 20000000000000.B"
+      mem50tb: "memory = 50000000000000.B"
+      mem100tb: "memory = 100000000000000.B"
+      mem200tb: "memory = 200000000000000.B"
+      mem500tb: "memory = 500000000000000.B"
+      mem1gib: "memory = 1073741824.B"
+      mem2gib: "memory = 2147483648.B"
+      mem4gib: "memory = 4294967296.B"
+      mem8gib: "memory = 8589934592.B"
+      mem16gib: "memory = 17179869184.B"
+      mem32gib: "memory = 34359738368.B"
+      mem64gib: "memory = 68719476736.B"
+      mem128gib: "memory = 137438953472.B"
+      mem256gib: "memory = 274877906944.B"
+      mem512gib: "memory = 549755813888.B"
+      mem1tib: "memory = 1099511627776.B"
+      mem2tib: "memory = 2199023255552.B"
+      mem4tib: "memory = 4398046511104.B"
+      mem8tib: "memory = 8796093022208.B"
+      mem16tib: "memory = 17592186044416.B"
+      mem32tib: "memory = 35184372088832.B"
+      mem64tib: "memory = 70368744177664.B"
+      mem128tib: "memory = 140737488355328.B"
+      mem256tib: "memory = 281474976710656.B"
+      mem512tib: "memory = 562949953421312.B"
+      cpu1: "cpus = 1"
+      cpu2: "cpus = 2"
+      cpu5: "cpus = 5"
+      cpu10: "cpus = 10"
+      cpu20: "cpus = 20"
+      cpu50: "cpus = 50"
+      cpu100: "cpus = 100"
+      cpu200: "cpus = 200"
+      cpu500: "cpus = 500"
+      cpu1000: "cpus = 1000"
+    script:
+    - "includeConfig(\"nextflow_labels.config\")"
+  debug: false
+  container: "docker"
+engines:
+- type: "native"
+  id: "native"
+build_info:
+  config: "src/single_cell/process_integrate_annotate/config.vsh.yaml"
+  runner: "nextflow"
+  engine: "native"
+  output: "target/nextflow/single_cell/process_integrate_annotate"
+  executable: "target/nextflow/single_cell/process_integrate_annotate/main.nf"
+  viash_version: "0.9.7"
+  git_commit: "50682718d61f38677e46dde857fb948846ec6cf6"
+  git_remote: "https://github.com/openpipelines-bio/openpipeline_composed"
+  dependencies:
+  - "target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples"
+  - "target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist"
+  - "target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches"
+  - "target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden"
+  - "target/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden"
+package_config:
+  name: "openpipeline_composed"
+  version: "set-public-scope"
+  info:
+    test_resources:
+    - type: "s3"
+      path: "s3://openpipelines-bio/openpipeline_incubator/resources_test"
+      dest: "resources_test"
+  repositories:
+  - type: "vsh"
+    name: "openpipeline"
+    repo: "openpipeline"
+    tag: "v4.0.4"
+  - type: "vsh"
+    name: "openpipeline_qc"
+    repo: "openpipeline_qc"
+    tag: "v0.2.2"
+  - type: "vsh"
+    name: "biobox"
+    repo: "biobox"
+    tag: "v0.4.2"
+  viash_version: "0.9.7"
+  source: "src"
+  target: "target"
+  config_mods:
+  - ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag\
+    \ := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n\
+    .runners[.type == 'nextflow'].config.script := 'includeConfig(\"nextflow_labels.config\"\
+    )'\n"
+  - ".engines += { type: \"native\" }"
+  - ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'"
+  - ".engines[.type == 'docker'].target_tag := 'set-public-scope'"
+  organization: "vsh"
+  links:
+    repository: "https://github.com/openpipelines-bio/openpipeline_composed"
+    docker_registry: "ghcr.io"
diff --git a/target/nextflow/single_cell/process_integrate_annotate/main.nf b/target/nextflow/single_cell/process_integrate_annotate/main.nf
new file mode 100644
index 0000000..e6b1bb7
--- /dev/null
+++ b/target/nextflow/single_cell/process_integrate_annotate/main.nf
@@ -0,0 +1,4407 @@
+// process_integrate_annotate set-public-scope
+// 
+// This wrapper script is auto-generated by viash 0.9.7 and is thus a derivative
+// work thereof. This software comes with ABSOLUTELY NO WARRANTY from Data
+// Intuitive.
+// 
+// The component may contain files which fall under a different license. The
+// authors of this component should specify the license in the header of such
+// files, or include a separate license file detailing the licenses of all included
+// files.
+// 
+// Component authors:
+//  * Dorien Roosen (author, maintainer)
+//  * Weiwei Schultz (contributor)
+
+////////////////////////////
+// VDSL3 helper functions //
+////////////////////////////
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_checkArgumentType.nf'
+class UnexpectedArgumentTypeException extends Exception {
+  String errorIdentifier
+  String stage
+  String plainName
+  String expectedClass
+  String foundClass
+  
+  // ${key ? " in module '$key'" : ""}${id ? " id '$id'" : ""}
+  UnexpectedArgumentTypeException(String errorIdentifier, String stage, String plainName, String expectedClass, String foundClass) {
+    super("Error${errorIdentifier ? " $errorIdentifier" : ""}:${stage ? " $stage" : "" } argument '${plainName}' has the wrong type. " +
+      "Expected type: ${expectedClass}. Found type: ${foundClass}")
+    this.errorIdentifier = errorIdentifier
+    this.stage = stage
+    this.plainName = plainName
+    this.expectedClass = expectedClass
+    this.foundClass = foundClass
+  }
+}
+
+/**
+  * Checks if the given value is of the expected type. If not, an exception is thrown.
+  *
+  * @param stage The stage of the argument (input or output)
+  * @param par The parameter definition
+  * @param value The value to check
+  * @param errorIdentifier The identifier to use in the error message
+  * @return The value, if it is of the expected type
+  * @throws UnexpectedArgumentTypeException If the value is not of the expected type
+*/
+def _checkArgumentType(String stage, Map par, Object value, String errorIdentifier) {
+  // expectedClass will only be != null if value is not of the expected type
+  def expectedClass = null
+  def foundClass = null
+  
+  // todo: split if need be
+  
+  if (!par.required && value == null) {
+    expectedClass = null
+  } else if (par.multiple) {
+    if (value !instanceof Collection) {
+      value = [value]
+    }
+    
+    // split strings
+    value = value.collectMany{ val ->
+      if (val instanceof String) {
+        // collect() to ensure that the result is a List and not simply an array
+        val.split(par.multiple_sep).collect()
+      } else {
+        [val]
+      }
+    }
+
+    // process globs
+    if (par.type == "file" && par.direction == "input") {
+      value = value.collect{ it instanceof String ? file(it, hidden: true) : it }.flatten()
+    }
+
+    // check types of elements in list
+    try {
+      value = value.collect { listVal ->
+        _checkArgumentType(stage, par + [multiple: false], listVal, errorIdentifier)
+      }
+    } catch (UnexpectedArgumentTypeException e) {
+      expectedClass = "List[${e.expectedClass}]"
+      foundClass = "List[${e.foundClass}]"
+    }
+  } else if (par.type == "string") {
+    // cast to string if need be. only cast if the value is a GString
+    if (value instanceof GString) {
+      value = value as String
+    }
+    expectedClass = value instanceof String ? null : "String"
+  } else if (par.type == "integer") {
+    // cast to integer if need be
+    if (value !instanceof Integer) {
+      try {
+        value = value as Integer
+      } catch (NumberFormatException e) {
+        expectedClass = "Integer"
+      }
+    }
+  } else if (par.type == "long") {
+    // cast to long if need be
+    if (value !instanceof Long) {
+      try {
+        value = value as Long
+      } catch (NumberFormatException e) {
+        expectedClass = "Long"
+      }
+    }
+  } else if (par.type == "double") {
+    // cast to double if need be
+    if (value !instanceof Double) {
+      try {
+        value = value as Double
+      } catch (NumberFormatException e) {
+        expectedClass = "Double"
+      }
+    }
+  } else if (par.type == "float") {
+    // cast to float if need be
+    if (value !instanceof Float) {
+      try {
+        value = value as Float
+      } catch (NumberFormatException e) {
+        expectedClass = "Float"
+      }
+    }
+  } else if (par.type == "boolean" | par.type == "boolean_true" | par.type == "boolean_false") {
+    // cast to boolean if need be
+    if (value !instanceof Boolean) {
+      try {
+        value = value as Boolean
+      } catch (Exception e) {
+        expectedClass = "Boolean"
+      }
+    }
+  } else if (par.type == "file" && (par.direction == "input" || stage == "output")) {
+    // cast to path if need be
+    if (value instanceof String) {
+      value = file(value, hidden: true)
+    }
+    if (value instanceof File) {
+      value = value.toPath()
+    }
+    expectedClass = value instanceof Path ? null : "Path"
+  } else if (par.type == "file" && stage == "input" && par.direction == "output") {
+    // cast to string if need be
+    if (value !instanceof String) {
+      try {
+        value = value as String
+      } catch (Exception e) {
+        expectedClass = "String"
+      }
+    }
+  } else {
+    // didn't find a match for par.type
+    expectedClass = par.type
+  }
+
+  if (expectedClass != null) {
+    if (foundClass == null) {
+      foundClass = value.getClass().getName()
+    }
+    throw new UnexpectedArgumentTypeException(errorIdentifier, stage, par.plainName, expectedClass, foundClass)
+  }
+  
+  return value
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processInputValues.nf'
+Map _processInputValues(Map inputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.required && arg.direction == "input") {
+        assert inputs.containsKey(arg.plainName) && inputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required input argument '${arg.plainName}' is missing"
+      }
+    }
+
+    inputs = inputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid input argument"
+
+      value = _checkArgumentType("input", par, value, "in module '$key' id '$id'")
+
+      [ name, value ]
+    }
+  }
+  return inputs
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/arguments/_processOutputValues.nf'
+Map _checkValidOutputArgument(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    outputs = outputs.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && it.direction == "output" }
+      assert par != null : "Error in module '${key}' id '${id}': '${name}' is not a valid output argument"
+      
+      value = _checkArgumentType("output", par, value, "in module '$key' id '$id'")
+      
+      [ name, value ]
+    }
+  }
+  return outputs
+}
+
+void _checkAllRequiredOuputsPresent(Map outputs, Map config, String id, String key) {
+  if (!workflow.stubRun) {
+    config.allArguments.each { arg ->
+      if (arg.direction == "output" && arg.required) {
+        assert outputs.containsKey(arg.plainName) && outputs.get(arg.plainName) != null : 
+          "Error in module '${key}' id '${id}': required output argument '${arg.plainName}' is missing"
+      }
+    }
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/IDChecker.nf'
+class IDChecker {
+  final def items = [] as Set
+
+  @groovy.transform.WithWriteLock
+  boolean observe(String item) {
+    if (items.contains(item)) {
+      return false
+    } else {
+      items << item
+      return true
+    }
+  }
+
+  @groovy.transform.WithReadLock
+  boolean contains(String item) {
+    return items.contains(item)
+  }
+
+  @groovy.transform.WithReadLock
+  Set getItems() {
+    return items.clone()
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_checkUniqueIds.nf'
+
+/**
+ * Check if the ids are unique across parameter sets
+ *
+ * @param parameterSets a list of parameter sets.
+ */
+private void _checkUniqueIds(List<Tuple2<String, Map<String, Object>>> parameterSets) {
+  def ppIds = parameterSets.collect{it[0]}
+  assert ppIds.size() == ppIds.unique().size() : "All argument sets should have unique ids. Detected ids: $ppIds"
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_getChild.nf'
+
+// helper functions for reading params from file //
+def _getChild(parent, child) {
+  if (child.contains("://") || java.nio.file.Paths.get(child).isAbsolute()) {
+    child
+  } else {
+    def parentAbsolute = java.nio.file.Paths.get(parent).toAbsolutePath().toString()
+    parentAbsolute.replaceAll('/[^/]*$', "/") + child
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_parseParamList.nf'
+/**
+  * Figure out the param list format based on the file extension
+  *
+  * @param param_list A String containing the path to the parameter list file.
+  *
+  * @return A String containing the format of the parameter list file.
+  */
+def _paramListGuessFormat(param_list) {
+  if (param_list !instanceof String) {
+    "asis"
+  } else if (param_list.endsWith(".csv")) {
+    "csv"
+  } else if (param_list.endsWith(".json") || param_list.endsWith(".jsn")) {
+    "json"
+  } else if (param_list.endsWith(".yaml") || param_list.endsWith(".yml")) {
+    "yaml"
+  } else {
+    "yaml_blob"
+  }
+}
+
+
+/**
+  * Read the param list
+  * 
+  * @param param_list One of the following:
+  *   - A String containing the path to the parameter list file (csv, json or yaml),
+  *   - A yaml blob of a list of maps (yaml_blob),
+  *   - Or a groovy list of maps (asis).
+  * @param config A Map of the Viash configuration.
+  * 
+  * @return A List of Maps containing the parameters.
+  */
+def _parseParamList(param_list, Map config) {
+  // first determine format by extension
+  def paramListFormat = _paramListGuessFormat(param_list)
+
+  def paramListPath = (paramListFormat != "asis" && paramListFormat != "yaml_blob") ?
+    file(param_list, hidden: true) :
+    null
+
+  // get the correct parser function for the detected params_list format
+  def paramSets = []
+  if (paramListFormat == "asis") {
+    paramSets = param_list
+  } else if (paramListFormat == "yaml_blob") {
+    paramSets = readYamlBlob(param_list)
+  } else if (paramListFormat == "yaml") {
+    paramSets = readYaml(paramListPath)
+  } else if (paramListFormat == "json") {
+    paramSets = readJson(paramListPath)
+  } else if (paramListFormat == "csv") {
+    paramSets = readCsv(paramListPath)
+  } else {
+    error "Format of provided --param_list not recognised.\n" +
+    "Found: '$paramListFormat'.\n" +
+    "Expected: a csv file, a json file, a yaml file,\n" +
+    "a yaml blob or a groovy list of maps."
+  }
+
+  // data checks
+  assert paramSets instanceof List: "--param_list should contain a list of maps"
+  for (value in paramSets) {
+    assert value instanceof Map: "--param_list should contain a list of maps"
+  }
+
+  // id is argument
+  def idIsArgument = config.allArguments.any{it.plainName == "id"}
+
+  // Reformat from List<Map> to List<Tuple2<String, Map>> by adding the ID as first element of a Tuple2
+  paramSets = paramSets.collect({ data ->
+    def id = data.id
+    if (!idIsArgument) {
+      data = data.findAll{k, v -> k != "id"}
+    }
+    [id, data]
+  })
+
+  // Split parameters with 'multiple: true'
+  paramSets = paramSets.collect({ id, data ->
+    data = _splitParams(data, config)
+    [id, data]
+  })
+  
+  // The paths of input files inside a param_list file may have been specified relatively to the
+  // location of the param_list file. These paths must be made absolute.
+  if (paramListPath) {
+    paramSets = paramSets.collect({ id, data ->
+      def new_data = data.collectEntries{ parName, parValue ->
+        def par = config.allArguments.find{it.plainName == parName}
+        if (par && par.type == "file" && par.direction == "input") {
+          if (parValue instanceof Collection) {
+            parValue = parValue.collectMany{path -> 
+              def x = _resolveSiblingIfNotAbsolute(path, paramListPath)
+              x instanceof Collection ? x : [x]
+            }
+          } else {
+            parValue = _resolveSiblingIfNotAbsolute(parValue, paramListPath) 
+          }
+        }
+        [parName, parValue]
+      }
+      [id, new_data]
+    })
+  }
+
+  return paramSets
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/_splitParams.nf'
+/**
+ * Split parameters for arguments that accept multiple values using their separator
+ *
+ * @param paramList A Map containing parameters to split.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ *
+ * @return A Map of parameters where the parameter values have been split into a list using
+ *         their seperator.
+ */
+Map<String, Object> _splitParams(Map<String, Object> parValues, Map config){
+  def parsedParamValues = parValues.collectEntries { parName, parValue ->
+    def parameterSettings = config.allArguments.find({it.plainName == parName})
+
+    if (!parameterSettings) {
+      // if argument is not found, do not alter 
+      return [parName, parValue]
+    }
+    if (parameterSettings.multiple) { // Check if parameter can accept multiple values
+      if (parValue instanceof Collection) {
+        parValue = parValue.collect{it instanceof String ? it.split(parameterSettings.multiple_sep) : it }
+      } else if (parValue instanceof String) {
+        parValue = parValue.split(parameterSettings.multiple_sep)
+      } else if (parValue == null) {
+        parValue = []
+      } else {
+        parValue = [ parValue ]
+      }
+      parValue = parValue.flatten()
+    }
+    // For all parameters check if multiple values are only passed for
+    // arguments that allow it. Quietly simplify lists of length 1.
+    if (!parameterSettings.multiple && parValue instanceof Collection) {
+      assert parValue.size() == 1 : 
+      "Error: argument ${parName} has too many values.\n" +
+      "  Expected amount: 1. Found: ${parValue.size()}"
+      parValue = parValue[0]
+    }
+    [parName, parValue]
+  }
+  return parsedParamValues
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/channelFromParams.nf'
+/**
+ * Parse nextflow parameters based on settings defined in a viash config.
+ * Return a list of parameter sets, each parameter set corresponding to 
+ * an event in a nextflow channel. The output from this function can be used
+ * with Channel.fromList to create a nextflow channel with Vdsl3 formatted 
+ * events.
+ *
+ * This function performs:
+ *   - A filtering of the params which can be found in the config file.
+ *   - Process the params_list argument which allows a user to to initialise 
+ *     a Vsdl3 channel with multiple parameter sets. Possible formats are 
+ *     csv, json, yaml, or simply a yaml_blob. A csv should have column names 
+ *     which correspond to the different arguments of this pipeline. A json or a yaml
+ *     file should be a list of maps, each of which has keys corresponding to the
+ *     arguments of the pipeline. A yaml blob can also be passed directly as a parameter.
+ *     When passing a csv, json or yaml, relative path names are relativized to the
+ *     location of the parameter file.
+ *   - Combine the parameter sets into a vdsl3 Channel.
+ *
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A list of parameters with the first element of the event being
+ *         the event ID and the second element containing a map of the parsed parameters.
+ */
+ 
+private List<Tuple2<String, Map<String, Object>>> _paramsToParamSets(Map params, Map config){
+  // todo: fetch key from run args
+  def key_ = config.name
+  
+  /* parse regular parameters (not in param_list)  */
+  /*************************************************/
+  def globalParams = config.allArguments
+    .findAll { params.containsKey(it.plainName) }
+    .collectEntries { [ it.plainName, params[it.plainName] ] }
+  def globalID = params.get("id", null)
+
+  /* process params_list arguments */
+  /*********************************/
+  def paramList = params.containsKey("param_list") && params.param_list != null ?
+    params.param_list : []
+  // if (paramList instanceof String) {
+  //   paramList = [paramList]
+  // }
+  // def paramSets = paramList.collectMany{ _parseParamList(it, config) }
+  // TODO: be able to process param_list when it is a list of strings
+  def paramSets = _parseParamList(paramList, config)
+  if (paramSets.isEmpty()) {
+    paramSets = [[null, [:]]]
+  }
+
+  /* combine arguments into channel */
+  /**********************************/
+  def processedParams = paramSets.indexed().collect{ index, tup ->
+    // Process ID
+    def id = tup[0] ?: globalID
+  
+    if (workflow.stubRun && !id) {
+      // if stub run, explicitly add an id if missing
+      id = "stub${index}"
+    }
+    assert id != null: "Each parameter set should have at least an 'id'"
+
+    // Process params
+    def parValues = globalParams + tup[1]
+    // // Remove parameters which are null, if the default is also null
+    // parValues = parValues.collectEntries{paramName, paramValue ->
+    //   parameterSettings = config.functionality.allArguments.find({it.plainName == paramName})
+    //   if ( paramValue != null || parameterSettings.get("default", null) != null ) {
+    //     [paramName, paramValue]
+    //   }
+    // }
+    parValues = parValues.collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      assert par != null : "Error in module '${key_}' id '${id}': '${name}' is not a valid input argument"
+
+      if (par == null) {
+        return [:]
+      }
+      value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+
+      [ name, value ]
+    }
+
+    [id, parValues]
+  }
+
+  // Check if ids (first element of each list) is unique
+  _checkUniqueIds(processedParams)
+  return processedParams
+}
+
+/**
+ * Parse nextflow parameters based on settings defined in a viash config 
+ * and return a nextflow channel.
+ * 
+ * @param params Input parameters. Can optionaly contain a 'param_list' key that
+ *               provides a list of arguments that can be split up into multiple events
+ *               in the output channel possible formats of param_lists are: a csv file, 
+ *               json file, a yaml file or a yaml blob. Each parameters set (event) must
+ *               have a unique ID.
+ * @param config A Map of the Viash configuration. This Map can be generated from the config file
+ *               using the readConfig() function.
+ * 
+ * @return A nextflow Channel with events. Events are formatted as a tuple that contains 
+ *         first contains the ID of the event and as second element holds a parameter map.
+ *       
+ *
+ */
+def channelFromParams(Map params, Map config) {
+  def processedParams = _paramsToParamSets(params, config)
+  return Channel.fromList(processedParams)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/checkUniqueIds.nf'
+def checkUniqueIds(Map args) {
+  def stopOnError = args.stopOnError == null ? args.stopOnError : true
+
+  def idChecker = new IDChecker()
+
+  return filter { tup ->
+    if (!idChecker.observe(tup[0])) {
+      if (stopOnError) {
+        error "Duplicate id: ${tup[0]}"
+      } else {
+        log.warn "Duplicate id: ${tup[0]}, removing duplicate entry"
+        return false
+      }
+    }
+    return true
+  }
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/preprocessInputs.nf'
+// This helper file will be deprecated soon
+preprocessInputsDeprecationWarningPrinted = false
+
+def preprocessInputsDeprecationWarning() {
+  if (!preprocessInputsDeprecationWarningPrinted) {
+    preprocessInputsDeprecationWarningPrinted = true
+    System.err.println("Warning: preprocessInputs() is deprecated and will be removed in Viash 0.9.0.")
+  }
+}
+
+/**
+ * Generate a nextflow Workflow that allows processing a channel of 
+ * Vdsl3 formatted events and apply a Viash config to them:
+ *    - Gather default parameters from the Viash config and make 
+ *      sure that they are correctly formatted (see applyConfig method).
+ *    - Format the input parameters (also using the applyConfig method).
+ *    - Apply the default parameter to the input parameters.
+ *    - Do some assertions:
+ *        ~ Check if the event IDs in the channel are unique.
+ * 
+ * The events in the channel are formatted as tuples, with the 
+ * first element of the tuples being a unique id of the parameter set, 
+ * and the second element containg the the parameters themselves.
+ * Optional extra elements of the tuples will be passed to the output as is.
+ *
+ * @param args A map that must contain a 'config' key that points
+ *              to a parsed config (see readConfig()). Optionally, a
+ *              'key' key can be provided which can be used to create a unique
+ *              name for the workflow process.
+ *
+ * @return A workflow that allows processing a channel of Vdsl3 formatted events
+ * and apply a Viash config to them.
+ */
+def preprocessInputs(Map args) {
+  preprocessInputsDeprecationWarning()
+
+  def config = args.config
+  assert config instanceof Map : 
+    "Error in preprocessInputs: config must be a map. " +
+    "Expected class: Map. Found: config.getClass() is ${config.getClass()}"
+  def key_ = args.key ?: config.name
+
+  // Get different parameter types (used throughout this function)
+  def defaultArgs = config.allArguments
+    .findAll { it.containsKey("default") }
+    .collectEntries { [ it.plainName, it.default ] }
+
+  map { tup ->
+    def id = tup[0]
+    def data = tup[1]
+    def passthrough = tup.drop(2)
+
+    def new_data = (defaultArgs + data).collectEntries { name, value ->
+      def par = config.allArguments.find { it.plainName == name && (it.direction == "input" || it.type == "file") }
+      
+      if (par != null) {
+        value = _checkArgumentType("input", par, value, "in module '$key_' id '$id'")
+      }
+
+      [ name, value ]
+    }
+
+    [ id, new_data ] + passthrough
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runComponents.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component config.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component config.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component config.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runComponents(Map args) {
+  log.warn("runComponents is deprecated, use runEach instead")
+  assert args.components: "runComponents should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runComponents"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def id_ = args.id
+
+  workflow runComponentsWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def comp_config = comp_.config
+
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_config)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          // def new_id = id_(tup[0], tup[1], comp_config)
+          def new_id = tup[0]
+          if (id_ instanceof String) {
+            new_id = id_
+          } else if (id_ instanceof Closure) {
+            new_id = id_(new_id, tup[1], comp_config)
+          }
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def data_ch = id_ch | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_config)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_config)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+      
+      post_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runComponentsWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/runEach.nf'
+/**
+ * Run a list of components on a stream of data.
+ * 
+ * @param components: list of Viash VDSL3 modules to run
+ * @param fromState: a closure, a map or a list of keys to extract from the input data.
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param toState: a closure, a map or a list of keys to extract from the output data
+ *   If a closure, it will be called with the id, the output data, the old state and the component itself.
+ * @param filter: filter function to apply to the input.
+ *   It will be called with the id, the data and the component itself.
+ * @param id: id to use for the output data
+ *   If a closure, it will be called with the id, the data and the component itself.
+ * @param auto: auto options to pass to the components
+ *
+ * @return: a workflow that runs the components
+ **/
+def runEach(Map args) {
+  assert args.components: "runEach should be passed a list of components to run"
+
+  def components_ = args.components
+  if (components_ !instanceof List) {
+    components_ = [ components_ ]
+  }
+  assert components_.size() > 0: "pass at least one component to runEach"
+
+  def fromState_ = args.fromState
+  def toState_ = args.toState
+  def filter_ = args.filter
+  def runIf_ = args.runIf
+  def id_ = args.id
+
+  assert !runIf_ || runIf_ instanceof Closure: "runEach: must pass a Closure to runIf."
+
+  workflow runEachWf {
+    take: input_ch
+    main:
+
+    // generate one channel per method
+    out_chs = components_.collect{ comp_ ->
+      def filter_ch = filter_
+        ? input_ch | filter{tup ->
+          filter_(tup[0], tup[1], comp_)
+        }
+        : input_ch
+      def id_ch = id_
+        ? filter_ch | map{tup ->
+          def new_id = id_
+          if (new_id instanceof Closure) {
+            new_id = new_id(tup[0], tup[1], comp_)
+          }
+          assert new_id instanceof String : "Error in runEach: id should be a String or a Closure that returns a String. Expected: id instanceof String. Found: ${new_id.getClass()}"
+          [new_id] + tup.drop(1)
+        }
+        : filter_ch
+      def chPassthrough = null
+      def chRun = null
+      if (runIf_) {
+        def idRunIfBranch = id_ch.branch{ tup ->
+          run: runIf_(tup[0], tup[1], comp_)
+          passthrough: true
+        }
+        chPassthrough = idRunIfBranch.passthrough
+        chRun = idRunIfBranch.run
+      } else {
+        chRun = id_ch
+        chPassthrough = Channel.empty()
+      }
+      def data_ch = chRun | map{tup ->
+          def new_data = tup[1]
+          if (fromState_ instanceof Map) {
+            new_data = fromState_.collectEntries{ key0, key1 ->
+              [key0, new_data[key1]]
+            }
+          } else if (fromState_ instanceof List) {
+            new_data = fromState_.collectEntries{ key ->
+              [key, new_data[key]]
+            }
+          } else if (fromState_ instanceof Closure) {
+            new_data = fromState_(tup[0], new_data, comp_)
+          }
+          tup.take(1) + [new_data] + tup.drop(1)
+        }
+      def out_ch = data_ch
+        | comp_.run(
+          auto: (args.auto ?: [:]) + [simplifyInput: false, simplifyOutput: false]
+        )
+      def post_ch = toState_
+        ? out_ch | map{tup ->
+          def output = tup[1]
+          def old_state = tup[2]
+          def new_state = null
+          if (toState_ instanceof Map) {
+            new_state = old_state + toState_.collectEntries{ key0, key1 ->
+              [key0, output[key1]]
+            }
+          } else if (toState_ instanceof List) {
+            new_state = old_state + toState_.collectEntries{ key ->
+              [key, output[key]]
+            }
+          } else if (toState_ instanceof Closure) {
+            new_state = toState_(tup[0], output, old_state, comp_)
+          }
+          [tup[0], new_state] + tup.drop(3)
+        }
+        : out_ch
+
+      def return_ch = post_ch
+        | concat(chPassthrough)
+      
+      return_ch
+    }
+
+    // mix all results
+    output_ch =
+      (out_chs.size == 1)
+        ? out_chs[0]
+        : out_chs[0].mix(*out_chs.drop(1))
+
+    emit: output_ch
+  }
+
+  return runEachWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/channel/safeJoin.nf'
+/**
+ * Join sourceChannel to targetChannel
+ * 
+ * This function joins the sourceChannel to the targetChannel. 
+ * However, each id in the targetChannel must be present in the
+ * sourceChannel. If _meta.join_id exists in the targetChannel, that is 
+ * used as an id instead. If the id doesn't match any id in the sourceChannel,
+ * an error is thrown.
+ */
+
+def safeJoin(targetChannel, sourceChannel, key) {
+  def sourceIDs = new IDChecker()
+
+  def sourceCheck = sourceChannel
+    | map { tup ->
+      sourceIDs.observe(tup[0])
+      tup
+    }
+  def targetCheck = targetChannel
+    | map { tup ->
+      def id = tup[0]
+      
+      if (!sourceIDs.contains(id)) {
+        error (
+          "Error in module '${key}' when merging output with original state.\n" +
+          "  Reason: output with id '${id}' could not be joined with source channel.\n" +
+          "    If the IDs in the output channel differ from the input channel,\n" + 
+          "    please set `tup[1]._meta.join_id to the original ID.\n" +
+          "  Original IDs in input channel: ['${sourceIDs.getItems().join("', '")}'].\n" + 
+          "  Unexpected ID in the output channel: '${id}'.\n" +
+          "  Example input event: [\"id\", [input: file(...)]],\n" +
+          "  Example output event: [\"newid\", [output: file(...), _meta: [join_id: \"id\"]]]"
+        )
+      }
+      // TODO: add link to our documentation on how to fix this
+
+      tup
+    }
+  
+  sourceCheck.cross(targetChannel)
+    | map{ left, right ->
+      right + left.drop(1)
+    }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/_processArgument.nf'
+def _processArgument(arg) {
+  arg.multiple = arg.multiple != null ? arg.multiple : false
+  arg.required = arg.required != null ? arg.required : false
+  arg.direction = arg.direction != null ? arg.direction : "input"
+  arg.multiple_sep = arg.multiple_sep != null ? arg.multiple_sep : ";"
+  arg.plainName = arg.name.replaceAll("^-*", "")
+
+  if (arg.type == "file") {
+    arg.must_exist = arg.must_exist != null ? arg.must_exist : true
+    arg.create_parent = arg.create_parent != null ? arg.create_parent : true
+  }
+
+  // add default values to output files which haven't already got a default
+  if (arg.type == "file" && arg.direction == "output" && arg.default == null) {
+    def mult = arg.multiple ? "_*" : ""
+    def extSearch = ""
+    if (arg.default != null) {
+      extSearch = arg.default
+    } else if (arg.example != null) {
+      extSearch = arg.example
+    }
+    if (extSearch instanceof List) {
+      extSearch = extSearch[0]
+    }
+    def extSearchResult = extSearch.find("\\.[^\\.]+\$")
+    def ext = extSearchResult != null ? extSearchResult : ""
+    arg.default = "\$id.\$key.${arg.plainName}${mult}${ext}"
+    if (arg.multiple) {
+      arg.default = [arg.default]
+    }
+  }
+
+  if (!arg.multiple) {
+    if (arg.default != null && arg.default instanceof List) {
+      arg.default = arg.default[0]
+    }
+    if (arg.example != null && arg.example instanceof List) {
+      arg.example = arg.example[0]
+    }
+  }
+
+  if (arg.type == "boolean_true") {
+    arg.default = false
+  }
+  if (arg.type == "boolean_false") {
+    arg.default = true
+  }
+
+  arg
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/addGlobalParams.nf'
+def addGlobalArguments(config) {
+  def localConfig = [
+    "argument_groups": [
+      [
+        "name": "Nextflow input-output arguments",
+        "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+        "arguments" : [
+          [
+            'name': '--publish_dir',
+            'required': true,
+            'type': 'string',
+            'description': 'Path to an output directory.',
+            'example': 'output/',
+            'multiple': false
+          ],
+          [
+            'name': '--param_list',
+            'required': false,
+            'type': 'string',
+            'description': '''Allows inputting multiple parameter sets to initialise a Nextflow channel. A `param_list` can either be a list of maps, a csv file, a json file, a yaml file, or simply a yaml blob.
+            |
+            |* A list of maps (as-is) where the keys of each map corresponds to the arguments of the pipeline. Example: in a `nextflow.config` file: `param_list: [ ['id': 'foo', 'input': 'foo.txt'], ['id': 'bar', 'input': 'bar.txt'] ]`.
+            |* A csv file should have column names which correspond to the different arguments of this pipeline. Example: `--param_list data.csv` with columns `id,input`.
+            |* A json or a yaml file should be a list of maps, each of which has keys corresponding to the arguments of the pipeline. Example: `--param_list data.json` with contents `[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]`.
+            |* A yaml blob can also be passed directly as a string. Example: `--param_list "[ {'id': 'foo', 'input': 'foo.txt'}, {'id': 'bar', 'input': 'bar.txt'} ]"`.
+            |
+            |When passing a csv, json or yaml file, relative path names are relativized to the location of the parameter file. No relativation is performed when `param_list` is a list of maps (as-is) or a yaml blob.'''.stripMargin(),
+            'example': 'my_params.yaml',
+            'multiple': false,
+            'hidden': true
+          ]
+          // TODO: allow multiple: true in param_list?
+          // TODO: allow to specify a --param_list_regex to filter the param_list?
+          // TODO: allow to specify a --param_list_from_state to remap entries in the param_list?
+        ]
+      ]
+    ]
+  ]
+
+  return processConfig(_mergeMap(config, localConfig))
+}
+
+def _mergeMap(Map lhs, Map rhs) {
+  return rhs.inject(lhs.clone()) { map, entry ->
+    if (map[entry.key] instanceof Map && entry.value instanceof Map) {
+      map[entry.key] = _mergeMap(map[entry.key], entry.value)
+    } else if (map[entry.key] instanceof Collection && entry.value instanceof Collection) {
+      map[entry.key] += entry.value
+    } else {
+      map[entry.key] = entry.value
+    }
+    return map
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/generateHelp.nf'
+def _generateArgumentHelp(param) {
+  // alternatives are not supported
+  // def names = param.alternatives ::: List(param.name)
+
+  def unnamedProps = [
+    ["required parameter", param.required],
+    ["multiple values allowed", param.multiple],
+    ["output", param.direction.toLowerCase() == "output"],
+    ["file must exist", param.type == "file" && param.must_exist]
+  ].findAll{it[1]}.collect{it[0]}
+  
+  def dflt = null
+  if (param.default != null) {
+    if (param.default instanceof List) {
+      dflt = param.default.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      dflt = param.default.toString()
+    }
+  }
+  def example = null
+  if (param.example != null) {
+    if (param.example instanceof List) {
+      example = param.example.join(param.multiple_sep != null ? param.multiple_sep : ", ")
+    } else {
+      example = param.example.toString()
+    }
+  }
+  def min = param.min?.toString()
+  def max = param.max?.toString()
+
+  def escapeChoice = { choice ->
+    def s1 = choice.replaceAll("\\n", "\\\\n")
+    def s2 = s1.replaceAll("\"", """\\\"""")
+    s2.contains(",") || s2 != choice ? "\"" + s2 + "\"" : s2
+  }
+  def choices = param.choices == null ? 
+    null : 
+    "[ " + param.choices.collect{escapeChoice(it.toString())}.join(", ") + " ]"
+
+  def namedPropsStr = [
+    ["type", ([param.type] + unnamedProps).join(", ")],
+    ["default", dflt],
+    ["example", example],
+    ["choices", choices],
+    ["min", min],
+    ["max", max]
+  ]
+    .findAll{it[1]}
+    .collect{"\n        " + it[0] + ": " + it[1].replaceAll("\n", "\\n")}
+    .join("")
+  
+  def descStr = param.description == null ?
+    "" :
+    _paragraphWrap("\n" + param.description.trim(), 80 - 8).join("\n        ")
+  
+  "\n    --" + param.plainName +
+    namedPropsStr +
+    descStr
+}
+
+// Based on Helper.generateHelp() in Helper.scala
+def _generateHelp(config) {
+  def fun = config
+
+  // PART 1: NAME AND VERSION
+  def nameStr = fun.name + 
+    (fun.version == null ? "" : " " + fun.version)
+
+  // PART 2: DESCRIPTION
+  def descrStr = fun.description == null ? 
+    "" :
+    "\n\n" + _paragraphWrap(fun.description.trim(), 80).join("\n")
+
+  // PART 3: Usage
+  def usageStr = fun.usage == null ? 
+    "" :
+    "\n\nUsage:\n" + fun.usage.trim()
+
+  // PART 4: Options
+  def argGroupStrs = fun.allArgumentGroups.collect{argGroup ->
+    def name = argGroup.name
+    def descriptionStr = argGroup.description == null ?
+      "" :
+      "\n    " + _paragraphWrap(argGroup.description.trim(), 80-4).join("\n    ") + "\n"
+    def arguments = argGroup.arguments.collect{arg -> 
+      arg instanceof String ? fun.allArguments.find{it.plainName == arg} : arg
+    }.findAll{it != null}
+    def argumentStrs = arguments.collect{param -> _generateArgumentHelp(param)}
+    
+    "\n\n$name:" +
+      descriptionStr +
+      argumentStrs.join("\n")
+  }
+
+  // FINAL: combine
+  def out = nameStr + 
+    descrStr +
+    usageStr + 
+    argGroupStrs.join("")
+
+  return out
+}
+
+// based on Format._paragraphWrap
+def _paragraphWrap(str, maxLength) {
+  def outLines = []
+  str.split("\n").each{par ->
+    def words = par.split("\\s").toList()
+
+    def word = null
+    def line = words.pop()
+    while(!words.isEmpty()) {
+      word = words.pop()
+      if (line.length() + word.length() + 1 <= maxLength) {
+        line = line + " " + word
+      } else {
+        outLines.add(line)
+        line = word
+      }
+    }
+    if (words.isEmpty()) {
+      outLines.add(line)
+    }
+  }
+  return outLines
+}
+
+def helpMessage(config) {
+  if (params.containsKey("help") && params.help) {
+    def mergedConfig = addGlobalArguments(config)
+    def helpStr = _generateHelp(mergedConfig)
+    println(helpStr)
+    exit 0
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/processConfig.nf'
+def processConfig(config) {
+  // set defaults for arguments
+  config.arguments = 
+    (config.arguments ?: []).collect{_processArgument(it)}
+
+  // set defaults for argument_group arguments
+  config.argument_groups =
+    (config.argument_groups ?: []).collect{grp ->
+      grp.arguments = (grp.arguments ?: []).collect{_processArgument(it)}
+      grp
+    }
+
+  // create combined arguments list
+  config.allArguments = 
+    config.arguments +
+    config.argument_groups.collectMany{it.arguments}
+
+  // add missing argument groups (based on Functionality::allArgumentGroups())
+  def argGroups = config.argument_groups
+  if (argGroups.any{it.name.toLowerCase() == "arguments"}) {
+    argGroups = argGroups.collect{ grp ->
+      if (grp.name.toLowerCase() == "arguments") {
+        grp = grp + [
+          arguments: grp.arguments + config.arguments
+        ]
+      }
+      grp
+    }
+  } else {
+    argGroups = argGroups + [
+      name: "Arguments",
+      arguments: config.arguments
+    ]
+  }
+  config.allArgumentGroups = argGroups
+
+  config
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/config/readConfig.nf'
+
+def readConfig(file) {
+  def config = readYaml(file ?: moduleDir.resolve("config.vsh.yaml"))
+  processConfig(config)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_resolveSiblingIfNotAbsolute.nf'
+/**
+  * Resolve a path relative to the current file.
+  * 
+  * @param str The path to resolve, as a String.
+  * @param parentPath The path to resolve relative to, as a Path.
+  *
+  * @return The path that may have been resovled, as a Path.
+  */
+def _resolveSiblingIfNotAbsolute(str, parentPath) {
+  if (str !instanceof String) {
+    return str
+  }
+  if (!_stringIsAbsolutePath(str)) {
+    return parentPath.resolveSibling(str)
+  } else {
+    return file(str, hidden: true)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/_stringIsAbsolutePath.nf'
+/**
+  * Check whether a path as a string is absolute.
+  *
+  * In the past, we tried using `file(., relative: true).isAbsolute()`,
+  * but the 'relative' option was added in 22.10.0.
+  *
+  * @param path The path to check, as a String.
+  *
+  * @return Whether the path is absolute, as a boolean.
+  */
+def _stringIsAbsolutePath(path) {
+  def _resolve_URL_PROTOCOL = ~/^([a-zA-Z][a-zA-Z0-9]*:)?\\/.+/
+
+  assert path instanceof String
+  return _resolve_URL_PROTOCOL.matcher(path).matches()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/collectTraces.nf'
+class CustomTraceObserver implements nextflow.trace.TraceObserver {
+  List traces
+
+  CustomTraceObserver(List traces) {
+    this.traces = traces
+  }
+
+  @Override
+  void onProcessComplete(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+
+  @Override
+  void onProcessCached(nextflow.processor.TaskHandler handler, nextflow.trace.TraceRecord trace) {
+    def trace2 = trace.store.clone()
+    trace2.script = null
+    traces.add(trace2)
+  }
+}
+
+def collectTraces() {
+  def traces = Collections.synchronizedList([])
+
+  // add custom trace observer which stores traces in the traces object
+  session.observers.add(new CustomTraceObserver(traces))
+
+  traces
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/deepClone.nf'
+/**
+  * Performs a deep clone of the given object.
+  * @param x an object
+  */
+def deepClone(x) {
+  iterateMap(x, {it instanceof Cloneable ? it.clone() : it})
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getPublishDir.nf'
+def getPublishDir() {
+  return params.containsKey("publish_dir") ? params.publish_dir : 
+    params.containsKey("publishDir") ? params.publishDir : 
+    null
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/getRootDir.nf'
+
+// Recurse upwards until we find a '.build.yaml' file
+def _findBuildYamlFile(pathPossiblySymlink) {
+  def path = pathPossiblySymlink.toRealPath()
+  def child = path.resolve(".build.yaml")
+  if (java.nio.file.Files.isDirectory(path) && java.nio.file.Files.exists(child)) {
+    return child
+  } else {
+    def parent = path.getParent()
+    if (parent == null) {
+      return null
+    } else {
+      return _findBuildYamlFile(parent)
+    }
+  }
+}
+
+// get the root of the target folder
+def getRootDir() {
+  def dir = _findBuildYamlFile(meta.resources_dir)
+  assert dir != null: "Could not find .build.yaml in the folder structure"
+  dir.getParent()
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/iterateMap.nf'
+/**
+  * Recursively apply a function over the leaves of an object.
+  * @param obj The object to iterate over.
+  * @param fun The function to apply to each value.
+  * @return The object with the function applied to each value.
+  */
+def iterateMap(obj, fun) {
+  if (obj instanceof List && obj !instanceof String) {
+    return obj.collect{item ->
+      iterateMap(item, fun)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectEntries{key, item ->
+      [key.toString(), iterateMap(item, fun)]
+    }
+  } else {
+    return fun(obj)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/functions/niceView.nf'
+/**
+  * A view for printing the event of each channel as a YAML blob.
+  * This is useful for debugging.
+  */
+def niceView() {
+  workflow niceViewWf {
+    take: input
+    main:
+      output = input
+        | view{toYamlBlob(it)}
+    emit: output
+  }
+  return niceViewWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readCsv.nf'
+
+def readCsv(file_path) {
+  def output = []
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+
+  // todo: allow escaped quotes in string
+  // todo: allow single quotes?
+  def splitRegex = java.util.regex.Pattern.compile(''',(?=(?:[^"]*"[^"]*")*[^"]*$)''')
+  def removeQuote = java.util.regex.Pattern.compile('''"(.*)"''')
+
+  java.nio.file.Files.newBufferedReader(inputFile).withCloseable { br ->
+    def row = 0
+    def header = null
+    def line
+
+    while (header == null && (line = br.readLine()) != null) {
+      if (!line.startsWith("#")) {
+        header = splitRegex.split(line, -1).collect { field ->
+          def m = removeQuote.matcher(field)
+          m.find() ? m.replaceFirst('$1') : field
+        }
+      }
+      row++
+    }
+    assert header != null : "CSV file should contain a header"
+
+    while ((line = br.readLine()) != null) {
+      row++
+      if (!line.startsWith("#")) {
+        def predata = splitRegex.split(line, -1)
+        def data = predata.collect { field ->
+          if (field == "") {
+            return null
+          }
+          def m = removeQuote.matcher(field)
+          if (m.find()) {
+            return m.replaceFirst('$1')
+          } else {
+            return field
+          }
+        }
+        assert header.size() == data.size() : "Row $row should contain the same number as fields as the header"
+
+        def dataMap = [header, data].transpose().collectEntries().findAll { it.value != null }
+        output.add(dataMap)
+      }
+    }
+  }
+
+  output
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJson.nf'
+def readJson(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parse(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readJsonBlob.nf'
+def readJsonBlob(str) {
+  def jsonSlurper = new groovy.json.JsonSlurper()
+  jsonSlurper.parseText(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readTaggedYaml.nf'
+// Custom constructor to modify how certain objects are parsed from YAML
+class CustomConstructor extends org.yaml.snakeyaml.constructor.Constructor {
+  Path root
+
+  class ConstructPath extends org.yaml.snakeyaml.constructor.AbstractConstruct {
+    public Object construct(org.yaml.snakeyaml.nodes.Node node) {
+      String filename = (String) constructScalar(node);
+      if (root != null) {
+        return root.resolve(filename);
+      }
+      return java.nio.file.Paths.get(filename);
+    }
+  }
+
+  CustomConstructor(org.yaml.snakeyaml.LoaderOptions options, Path root) {
+    super(options)
+    this.root = root
+    // Handling !file tag and parse it back to a File type
+    this.yamlConstructors.put(new org.yaml.snakeyaml.nodes.Tag("!file"), new ConstructPath())
+  }
+}
+
+def readTaggedYaml(Path path) {
+  def options = new org.yaml.snakeyaml.LoaderOptions()
+  def constructor = new CustomConstructor(options, path.getParent())
+  def yaml = new org.yaml.snakeyaml.Yaml(constructor)
+  return yaml.load(path.text)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYaml.nf'
+def readYaml(file_path) {
+  def inputFile = file_path !instanceof Path ? file(file_path, hidden: true) : file_path
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(inputFile)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/readYamlBlob.nf'
+def readYamlBlob(str) {
+  def yamlSlurper = new org.yaml.snakeyaml.Yaml()
+  yamlSlurper.load(str)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toJsonBlob.nf'
+String toJsonBlob(data) {
+  return groovy.json.JsonOutput.toJson(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toTaggedYamlBlob.nf'
+// Custom representer to modify how certain objects are represented in YAML
+class CustomRepresenter extends org.yaml.snakeyaml.representer.Representer {
+  Path relativizer
+
+  class RepresentPath implements org.yaml.snakeyaml.representer.Represent {
+    public String getFileName(Object obj) {
+      if (obj instanceof File) {
+        obj = ((File) obj).toPath();
+      }
+      if (obj !instanceof Path) {
+        throw new IllegalArgumentException("Object: " + obj + " is not a Path or File");
+      }
+      def path = (Path) obj;
+
+      if (relativizer != null) {
+        return relativizer.relativize(path).toString()
+      } else {
+        return path.toString()
+      }
+    }
+
+    public org.yaml.snakeyaml.nodes.Node representData(Object data) {
+      String filename = getFileName(data);
+      def tag = new org.yaml.snakeyaml.nodes.Tag("!file");
+      return representScalar(tag, filename);
+    }
+  }
+  CustomRepresenter(org.yaml.snakeyaml.DumperOptions options, Path relativizer) {
+    super(options)
+    this.relativizer = relativizer
+    this.representers.put(sun.nio.fs.UnixPath, new RepresentPath())
+    this.representers.put(Path, new RepresentPath())
+    this.representers.put(File, new RepresentPath())
+  }
+}
+
+String toTaggedYamlBlob(data) {
+  return toRelativeTaggedYamlBlob(data, null)
+}
+String toRelativeTaggedYamlBlob(data, Path relativizer) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  def representer = new CustomRepresenter(options, relativizer)
+  def yaml = new org.yaml.snakeyaml.Yaml(representer, options)
+  return yaml.dump(data)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/toYamlBlob.nf'
+String toYamlBlob(data) {
+  def options = new org.yaml.snakeyaml.DumperOptions()
+  options.setDefaultFlowStyle(org.yaml.snakeyaml.DumperOptions.FlowStyle.BLOCK)
+  options.setPrettyFlow(true)
+  def yaml = new org.yaml.snakeyaml.Yaml(options)
+  def cleanData = iterateMap(data, { it instanceof Path ? it.toString() : it })
+  return yaml.dump(cleanData)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeJson.nf'
+void writeJson(data, file) {
+  assert data: "writeJson: data should not be null"
+  assert file: "writeJson: file should not be null"
+  file.write(toJsonBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/readwrite/writeYaml.nf'
+void writeYaml(data, file) {
+  assert data: "writeYaml: data should not be null"
+  assert file: "writeYaml: file should not be null"
+  file.write(toYamlBlob(data))
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/findStates.nf'
+def findStates(Map params, Map config) {
+  def auto_config = deepClone(config)
+  def auto_params = deepClone(params)
+
+  auto_config = auto_config.clone()
+  // override arguments
+  auto_config.argument_groups = []
+  auto_config.arguments = [
+    [
+      type: "string",
+      name: "--id",
+      description: "A dummy identifier",
+      required: false
+    ],
+    [
+      type: "file",
+      name: "--input_states",
+      example: "/path/to/input/directory/**/state.yaml",
+      description: "Path to input directory containing the datasets to be integrated.",
+      required: true,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--filter",
+      example: "foo/.*/state.yaml",
+      description: "Regex to filter state files by path.",
+      required: false
+    ],
+    // to do: make this a yaml blob?
+    [
+      type: "string",
+      name: "--rename_keys",
+      example: ["newKey1:oldKey1", "newKey2:oldKey2"],
+      description: "Rename keys in the detected input files. This is useful if the input files do not match the set of input arguments of the workflow.",
+      required: false,
+      multiple: true,
+      multiple_sep: ";"
+    ],
+    [
+      type: "string",
+      name: "--settings",
+      example: '{"output_dataset": "dataset.h5ad", "k": 10}',
+      description: "Global arguments as a JSON glob to be passed to all components.",
+      required: false
+    ]
+  ]
+  if (!(auto_params.containsKey("id"))) {
+    auto_params["id"] = "auto"
+  }
+
+  // run auto config through processConfig once more
+  auto_config = processConfig(auto_config)
+
+  workflow findStatesWf {
+    helpMessage(auto_config)
+
+    output_ch = 
+      channelFromParams(auto_params, auto_config)
+        | flatMap { autoId, args ->
+
+          def globalSettings = args.settings ? readYamlBlob(args.settings) : [:]
+
+          // look for state files in input dir
+          def stateFiles = args.input_states
+
+          // filter state files by regex
+          if (args.filter) {
+            stateFiles = stateFiles.findAll{ stateFile ->
+              def stateFileStr = stateFile.toString()
+              def matcher = stateFileStr =~ args.filter
+              matcher.matches()}
+          }
+
+          // read in states
+          def states = stateFiles.collect { stateFile ->
+            def state_ = readTaggedYaml(stateFile)
+            [state_.id, state_]
+          }
+
+          // construct renameMap
+          if (args.rename_keys) {
+            def renameMap = args.rename_keys.collectEntries{renameString ->
+              def split = renameString.split(":")
+              assert split.size() == 2: "Argument 'rename_keys' should be of the form 'newKey:oldKey', or 'newKey:oldKey;newKey:oldKey' in case of multiple values"
+              split
+            }
+
+            // rename keys in state, only let states through which have all keys
+            // also add global settings
+            states = states.collectMany{id, state ->
+              def newState = [:]
+
+              for (key in renameMap.keySet()) {
+                def origKey = renameMap[key]
+                if (!(state.containsKey(origKey))) {
+                  return []
+                }
+                newState[key] = state[origKey]
+              }
+
+              [[id, globalSettings + newState]]
+            }
+          }
+
+          states
+        }
+    emit:
+    output_ch
+  }
+
+  return findStatesWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/joinStates.nf'
+def joinStates(Closure apply_) {
+  workflow joinStatesWf {
+    take: input_ch
+    main:
+    output_ch = input_ch
+      | toSortedList
+      | filter{ it.size() > 0 }
+      | map{ tups ->
+        def ids = tups.collect{it[0]}
+        def states = tups.collect{it[1]}
+        apply_(ids, states)
+      }
+
+    emit: output_ch
+  }
+  return joinStatesWf
+}
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishFiles.nf'
+def publishFiles(Map args) {
+  def key_ = args.get("key")
+
+  assert key_ != null : "publishFiles: key must be specified"
+  
+  workflow publishFilesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+          def inputFiles_ = inputoutputFilenames_[0]
+          def outputFilenames_ = inputoutputFilenames_[1]
+
+          [id_, inputFiles_, outputFilenames_]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesWf
+}
+
+process publishFilesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), path(inputFiles, stageAs: "_inputfile?/*"), val(outputFiles)
+  output:
+    tuple val(id), path{outputFiles}
+  script:
+  def copyCommands = [
+    inputFiles instanceof List ? inputFiles : [inputFiles],
+    outputFiles instanceof List ? outputFiles : [outputFiles]
+  ]
+    .transpose()
+    .collectMany{infile, outfile ->
+      def infileString = infile.toString()
+      def outfileString = outfile.toString()
+      if (infileString != outfileString) {
+        /* Trailing slashes are removed from both the source and destination arguments.
+           From source arguments, this is useful when a source argument may have a trailing slash
+           and specify a symbolic link to a directory. Without removing the slash, cp will dereference
+           the symbolic link. 
+           See https://www.gnu.org/software/coreutils/manual/html_node/Trailing-slashes.html#Trailing-slashes-1
+
+           For the destination path addding a trailing slash is a problem when publishing directories: 
+           it requires the destination directory to exist. This fails because we only create the parent
+           directories first.
+        */
+        def regexTrailingSlashes = ~/\/+$/
+        def infileNoTrailingSlash = infileString - regexTrailingSlashes
+        def outfileNoTrailingSlash = outfileString - regexTrailingSlashes
+        [
+          "[ -d \"\$(dirname '${outfileNoTrailingSlash}')\" ] || mkdir -p \"\$(dirname '${outfileNoTrailingSlash}')\"",
+          "cp -a '${infileNoTrailingSlash}' '${outfileNoTrailingSlash}'"
+        ]
+      } else {
+        // no need to copy if infile is the same as outfile
+        []
+      }
+    }
+  """
+  echo "Copying output files to destination folder"
+  ${copyCommands.join("\n  ")}
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishFilesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishFilesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishFilesByConfig: key must be specified"
+  
+  workflow publishFilesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+
+          // the processed state is a list of [key, value, inputPath, outputFilename] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - inputPath is a List[Path]
+          //   - outputFilename is a List[String]
+          //   - (inputPath, outputFilename) are the files that will be copied from src to dest (relative to the state.yaml)
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output OR multiple channels were emitted
+                // and the output was just not added to using the channel
+                // that is now being parsed
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[inputPath: [], outputFilename: []]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def inputPath = val instanceof File ? val.toPath() : val
+                    [inputPath: inputPath, outputFilename: filename_ix]
+                  }
+                  def transposedOutputs = ["inputPath", "outputFilename"].collectEntries{ key -> 
+                    [key, outputPerFile.collect{dic -> dic[key]}]
+                  }
+                  return [[key: plainName_] + transposedOutputs]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [[inputPath: [inputPath], outputFilename: [filename]]]
+                }
+              }
+          
+          def inputPaths = processedState.collectMany{it.inputPath}
+          def outputFilenames = processedState.collectMany{it.outputFilename}
+          
+
+          [id_, inputPaths, outputFilenames]
+        }
+        | publishFilesProc
+    emit: input_ch
+  }
+  return publishFilesSimpleWf
+}
+
+
+
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/publishStates.nf'
+def collectFiles(obj) {
+  if (obj instanceof java.io.File || obj instanceof Path)  {
+    return [obj]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.collectMany{item ->
+      collectFiles(item)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectFiles(item)
+    }
+  } else {
+    return []
+  }
+}
+
+/**
+ * Recurse through a state and collect all input files and their target output filenames.
+ * @param obj The state to recurse through.
+ * @param prefix The prefix to prepend to the output filenames.
+ */
+def collectInputOutputPaths(obj, prefix) {
+  if (obj instanceof File || obj instanceof Path)  {
+    def path = obj instanceof Path ? obj : obj.toPath()
+    def ext = path.getFileName().toString().find("\\.[^\\.]+\$") ?: ""
+    def newFilename = prefix + ext
+    return [[obj, newFilename]]
+  } else if (obj instanceof List && obj !instanceof String) {
+    return obj.withIndex().collectMany{item, ix ->
+      collectInputOutputPaths(item, prefix + "_" + ix)
+    }
+  } else if (obj instanceof Map) {
+    return obj.collectMany{key, item ->
+      collectInputOutputPaths(item, prefix + "." + key)
+    }
+  } else {
+    return []
+  }
+}
+
+def publishStates(Map args) {
+  def key_ = args.get("key")
+  def yamlTemplate_ = args.get("output_state", args.get("outputState", '$id.$key.state.yaml'))
+
+  assert key_ != null : "publishStates: key must be specified"
+  
+  workflow publishStatesWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1]
+
+          // the input files and the target output filenames
+          def inputoutputFilenames_ = collectInputOutputPaths(state_, id_ + "." + key_).transpose()
+
+          def yamlFilename = yamlTemplate_
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+
+            // TODO: do the pathnames in state_ match up with the outputFilenames_?
+
+          // convert state to yaml blob
+          def yamlBlob_ = toRelativeTaggedYamlBlob([id: id_] + state_, java.nio.file.Paths.get(yamlFilename))
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesWf
+}
+process publishStatesProc {
+  // todo: check publishpath?
+  publishDir path: "${getPublishDir()}/", mode: "copy"
+  tag "$id"
+  input:
+    tuple val(id), val(yamlBlob), val(yamlFile)
+  output:
+    tuple val(id), path{[yamlFile]}
+  script:
+  """
+  mkdir -p "\$(dirname '${yamlFile}')"
+  echo "Storing state as yaml"
+  cat > '${yamlFile}' << HERE
+${yamlBlob}
+HERE
+  """
+}
+
+
+// this assumes that the state contains no other values other than those specified in the config
+def publishStatesByConfig(Map args) {
+  def config = args.get("config")
+  assert config != null : "publishStatesByConfig: config must be specified"
+
+  def key_ = args.get("key", config.name)
+  assert key_ != null : "publishStatesByConfig: key must be specified"
+  
+  workflow publishStatesSimpleWf {
+    take: input_ch
+    main:
+      input_ch
+        | map { tup ->
+          def id_ = tup[0]
+          def state_ = tup[1] // e.g. [output: new File("myoutput.h5ad"), k: 10]
+          def origState_ = tup[2] // e.g. [output: '$id.$key.foo.h5ad']
+
+          // TODO: allow overriding the state.yaml template
+          // TODO TODO: if auto.publish == "state", add output_state as an argument
+          def yamlTemplate = params.containsKey("output_state") ? params.output_state : '$id.$key.state.yaml'
+          def yamlFilename = yamlTemplate
+            .replaceAll('\\$id', id_)
+            .replaceAll('\\$\\{id\\}', id_)
+            .replaceAll('\\$key', key_)
+            .replaceAll('\\$\\{key\\}', key_)
+          def yamlDir = java.nio.file.Paths.get(yamlFilename).getParent()
+
+          // the processed state is a list of [key, value] tuples, where
+          //   - key is a String
+          //   - value is any object that can be serialized to a Yaml (so a String/Integer/Long/Double/Boolean, a List, a Map, or a Path)
+          //   - (key, value) are the tuples that will be saved to the state.yaml file
+          def processedState =
+            config.allArguments
+              .findAll { it.direction == "output" }
+              .collectMany { par ->
+                def plainName_ = par.plainName
+                // if the state does not contain the key, it's an
+                // optional argument for which the component did 
+                // not generate any output
+                if (!state_.containsKey(plainName_)) {
+                  return []
+                }
+                def value = state_[plainName_]
+                // if the parameter is not a file, it should be stored
+                // in the state as-is, but is not something that needs 
+                // to be copied from the source path to the dest path
+                if (par.type != "file") {
+                  return [[key: plainName_, value: value]]
+                }
+                // if the orig state does not contain this filename,
+                // it's an optional argument for which the user specified
+                // that it should not be returned as a state
+                if (!origState_.containsKey(plainName_)) {
+                  return []
+                }
+                def filenameTemplate = origState_[plainName_]
+                // if the pararameter is multiple: true, fetch the template
+                if (par.multiple && filenameTemplate instanceof List) {
+                  filenameTemplate = filenameTemplate[0]
+                }
+                // instantiate the template
+                def filename = filenameTemplate
+                  .replaceAll('\\$id', id_)
+                  .replaceAll('\\$\\{id\\}', id_)
+                  .replaceAll('\\$key', key_)
+                  .replaceAll('\\$\\{key\\}', key_)
+                if (par.multiple) {
+                  // if the parameter is multiple: true, the filename
+                  // should contain a wildcard '*' that is replaced with
+                  // the index of the file
+                  assert filename.contains("*") : "Module '${key_}' id '${id_}': Multiple output files specified, but no wildcard '*' in the filename: ${filename}"
+                  def outputPerFile = value.withIndex().collect{ val, ix ->
+                    def filename_ix = filename.replace("*", ix.toString())
+                    def value_ = java.nio.file.Paths.get(filename_ix)
+                    // if id contains a slash
+                    if (yamlDir != null) {
+                      value_ = yamlDir.relativize(value_)
+                    }
+                    return value_
+                  }
+                  return [["key": plainName_, "value": outputPerFile]]
+                } else {
+                  def value_ = java.nio.file.Paths.get(filename)
+                  // if id contains a slash
+                  if (yamlDir != null) {
+                    value_ = yamlDir.relativize(value_)
+                  }
+                  def inputPath = value instanceof File ? value.toPath() : value
+                  return [["key": plainName_, value: value_]]
+                }
+              }
+              
+          
+          def updatedState_ = processedState.collectEntries{[it.key, it.value]}
+          
+          // convert state to yaml blob
+          def yamlBlob_ = toTaggedYamlBlob([id: id_] + updatedState_)
+
+          [id_, yamlBlob_, yamlFilename]
+        }
+        | publishStatesProc
+    emit: input_ch
+  }
+  return publishStatesSimpleWf
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/states/setState.nf'
+def setState(fun) {
+  assert fun instanceof Closure || fun instanceof Map || fun instanceof List :
+    "Error in setState: Expected process argument to be a Closure, a Map, or a List. Found: class ${fun.getClass()}"
+
+  // if fun is a List, convert to map
+  if (fun instanceof List) {
+    // check whether fun is a list[string]
+    assert fun.every{it instanceof CharSequence} : "Error in setState: argument is a List, but not all elements are Strings"
+    fun = fun.collectEntries{[it, it]}
+  }
+
+  // if fun is a map, convert to closure
+  if (fun instanceof Map) {
+    // check whether fun is a map[string, string]
+    assert fun.values().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all values are Strings"
+    assert fun.keySet().every{it instanceof CharSequence} : "Error in setState: argument is a Map, but not all keys are Strings"
+    def funMap = fun.clone()
+    // turn the map into a closure to be used later on
+    fun = { id_, state_ ->
+      assert state_ instanceof Map : "Error in setState: the state is not a Map"
+      funMap.collectMany{newkey, origkey ->
+        if (state_.containsKey(origkey)) {
+          [[newkey, state_[origkey]]]
+        } else {
+          []
+        }
+      }.collectEntries()
+    }
+  }
+
+  map { tup ->
+    def id = tup[0]
+    def state = tup[1]
+    def unfilteredState = fun(id, state)
+    def newState = unfilteredState.findAll{key, val -> val != null}
+    [id, newState] + tup.drop(2)
+  }
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processAuto.nf'
+// TODO: unit test processAuto
+def processAuto(Map auto) {
+  // remove null values
+  auto = auto.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = ["simplifyInput", "simplifyOutput", "transcript", "publish"]
+  def unexpectedKeys = auto.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty(), "unexpected keys in auto: '${unexpectedKeys.join("', '")}'"
+
+  // check auto.simplifyInput
+  assert auto.simplifyInput instanceof Boolean, "auto.simplifyInput must be a boolean"
+
+  // check auto.simplifyOutput
+  assert auto.simplifyOutput instanceof Boolean, "auto.simplifyOutput must be a boolean"
+
+  // check auto.transcript
+  assert auto.transcript instanceof Boolean, "auto.transcript must be a boolean"
+
+  // check auto.publish
+  assert auto.publish instanceof Boolean || auto.publish == "state", "auto.publish must be a boolean or 'state'"
+
+  return auto.subMap(expectedKeys)
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processDirectives.nf'
+def assertMapKeys(map, expectedKeys, requiredKeys, mapName) {
+  assert map instanceof Map : "Expected argument '$mapName' to be a Map. Found: class ${map.getClass()}"
+  map.forEach { key, val -> 
+    assert key in expectedKeys : "Unexpected key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+  requiredKeys.forEach { requiredKey -> 
+    assert map.containsKey(requiredKey) : "Missing required key '$key' in ${mapName ? mapName + " " : ""}map"
+  }
+}
+
+// TODO: unit test processDirectives
+def processDirectives(Map drctv) {
+  // remove null values
+  drctv = drctv.findAll{k, v -> v != null}
+
+  // check for unexpected keys
+  def expectedKeys = [
+    "accelerator", "afterScript", "beforeScript", "cache", "conda", "container", "containerOptions", "cpus", "disk", "echo", "errorStrategy", "executor", "machineType", "maxErrors", "maxForks", "maxRetries", "memory", "module", "penv", "pod", "publishDir", "queue", "label", "scratch", "storeDir", "stageInMode", "stageOutMode", "tag", "time"
+  ]
+  def unexpectedKeys = drctv.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Unexpected keys in process directive: '${unexpectedKeys.join("', '")}'"
+
+  /* DIRECTIVE accelerator
+    accepted examples:
+    - [ limit: 4, type: "nvidia-tesla-k80" ]
+  */
+  if (drctv.containsKey("accelerator")) {
+    assertMapKeys(drctv["accelerator"], ["type", "limit", "request", "runtime"], [], "accelerator")
+  }
+
+  /* DIRECTIVE afterScript
+    accepted examples:
+    - "source /cluster/bin/cleanup"
+  */
+  if (drctv.containsKey("afterScript")) {
+    assert drctv["afterScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE beforeScript
+    accepted examples:
+    - "source /cluster/bin/setup"
+  */
+  if (drctv.containsKey("beforeScript")) {
+    assert drctv["beforeScript"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cache
+    accepted examples:
+    - true
+    - false
+    - "deep"
+    - "lenient"
+  */
+  if (drctv.containsKey("cache")) {
+    assert drctv["cache"] instanceof CharSequence || drctv["cache"] instanceof Boolean
+    if (drctv["cache"] instanceof CharSequence) {
+      assert drctv["cache"] in ["deep", "lenient"] : "Unexpected value for cache"
+    }
+  }
+
+  /* DIRECTIVE conda
+    accepted examples:
+    - "bwa=0.7.15"
+    - "bwa=0.7.15 fastqc=0.11.5"
+    - ["bwa=0.7.15", "fastqc=0.11.5"]
+  */
+  if (drctv.containsKey("conda")) {
+    if (drctv["conda"] instanceof List) {
+      drctv["conda"] = drctv["conda"].join(" ")
+    }
+    assert drctv["conda"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE container
+    accepted examples:
+    - "foo/bar:tag"
+    - [ registry: "reg", image: "im", tag: "ta" ]
+      is transformed to "reg/im:ta"
+    - [ image: "im" ] 
+      is transformed to "im:latest"
+  */
+  if (drctv.containsKey("container")) {
+    assert drctv["container"] instanceof Map || drctv["container"] instanceof CharSequence
+    if (drctv["container"] instanceof Map) {
+      def m = drctv["container"]
+      assertMapKeys(m, [ "registry", "image", "tag" ], ["image"], "container")
+      def part1 = 
+        System.getenv('OVERRIDE_CONTAINER_REGISTRY') ? System.getenv('OVERRIDE_CONTAINER_REGISTRY') + "/" : 
+        params.containsKey("override_container_registry") ? params["override_container_registry"] + "/" : // todo: remove?
+        m.registry ? m.registry + "/" : 
+        ""
+      def part2 = m.image
+      def part3 = m.tag ? ":" + m.tag : ":latest"
+      drctv["container"] = part1 + part2 + part3
+    }
+  }
+
+  /* DIRECTIVE containerOptions
+    accepted examples:
+    - "--foo bar"
+    - ["--foo bar", "-f b"]
+  */
+  if (drctv.containsKey("containerOptions")) {
+    if (drctv["containerOptions"] instanceof List) {
+      drctv["containerOptions"] = drctv["containerOptions"].join(" ")
+    }
+    assert drctv["containerOptions"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE cpus
+    accepted examples:
+    - 1
+    - 10
+  */
+  if (drctv.containsKey("cpus")) {
+    assert drctv["cpus"] instanceof Integer
+  }
+
+  /* DIRECTIVE disk
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("disk")) {
+    assert drctv["disk"] instanceof CharSequence
+    // assert drctv["disk"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE echo
+    accepted examples:
+    - true
+    - false
+  */
+  if (drctv.containsKey("echo")) {
+    assert drctv["echo"] instanceof Boolean
+  }
+
+  /* DIRECTIVE errorStrategy
+    accepted examples:
+    - "terminate"
+    - "finish"
+  */
+  if (drctv.containsKey("errorStrategy")) {
+    assert drctv["errorStrategy"] instanceof CharSequence
+    assert drctv["errorStrategy"] in ["terminate", "finish", "ignore", "retry"] : "Unexpected value for errorStrategy"
+  }
+
+  /* DIRECTIVE executor
+    accepted examples:
+    - "local"
+    - "sge"
+  */
+  if (drctv.containsKey("executor")) {
+    assert drctv["executor"] instanceof CharSequence
+    assert drctv["executor"] in ["local", "sge", "uge", "lsf", "slurm", "pbs", "pbspro", "moab", "condor", "nqsii", "ignite", "k8s", "awsbatch", "google-pipelines"] : "Unexpected value for executor"
+  }
+
+  /* DIRECTIVE machineType
+    accepted examples:
+    - "n1-highmem-8"
+  */
+  if (drctv.containsKey("machineType")) {
+    assert drctv["machineType"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE maxErrors
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxErrors")) {
+    assert drctv["maxErrors"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxForks
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxForks")) {
+    assert drctv["maxForks"] instanceof Integer
+  }
+
+  /* DIRECTIVE maxRetries
+    accepted examples:
+    - 1
+    - 3
+  */
+  if (drctv.containsKey("maxRetries")) {
+    assert drctv["maxRetries"] instanceof Integer
+  }
+
+  /* DIRECTIVE memory
+    accepted examples:
+    - "1 GB"
+    - "2TB"
+    - "3.2KB"
+    - "10.B"
+  */
+  if (drctv.containsKey("memory")) {
+    assert drctv["memory"] instanceof CharSequence
+    // assert drctv["memory"].matches("[0-9]+(\\.[0-9]*)? *[KMGTPEZY]?B")
+    // ^ does not allow closures
+  }
+
+  /* DIRECTIVE module
+    accepted examples:
+    - "ncbi-blast/2.2.27"
+    - "ncbi-blast/2.2.27:t_coffee/10.0"
+    - ["ncbi-blast/2.2.27", "t_coffee/10.0"]
+  */
+  if (drctv.containsKey("module")) {
+    if (drctv["module"] instanceof List) {
+      drctv["module"] = drctv["module"].join(":")
+    }
+    assert drctv["module"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE penv
+    accepted examples:
+    - "smp"
+  */
+  if (drctv.containsKey("penv")) {
+    assert drctv["penv"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE pod
+    accepted examples:
+    - [ label: "key", value: "val" ]
+    - [ annotation: "key", value: "val" ]
+    - [ env: "key", value: "val" ]
+    - [ [label: "l", value: "v"], [env: "e", value: "v"]]
+  */
+  if (drctv.containsKey("pod")) {
+    if (drctv["pod"] instanceof Map) {
+      drctv["pod"] = [ drctv["pod"] ]
+    }
+    assert drctv["pod"] instanceof List
+    drctv["pod"].forEach { pod ->
+      assert pod instanceof Map
+      // TODO: should more checks be added?
+      // See https://www.nextflow.io/docs/latest/process.html?highlight=directives#pod
+      // e.g. does it contain 'label' and 'value', or 'annotation' and 'value', or ...?
+    }
+  }
+
+  /* DIRECTIVE publishDir
+    accepted examples:
+    - []
+    - [ [ path: "foo", enabled: true ], [ path: "bar", enabled: false ] ]
+    - "/path/to/dir" 
+      is transformed to [[ path: "/path/to/dir" ]]
+    - [ path: "/path/to/dir", mode: "cache" ]
+      is transformed to [[ path: "/path/to/dir", mode: "cache" ]]
+  */
+  // TODO: should we also look at params["publishDir"]?
+  if (drctv.containsKey("publishDir")) {
+    def pblsh = drctv["publishDir"]
+    
+    // check different options
+    assert pblsh instanceof List || pblsh instanceof Map || pblsh instanceof CharSequence
+    
+    // turn into list if not already so
+    // for some reason, 'if (!pblsh instanceof List) pblsh = [ pblsh ]' doesn't work.
+    pblsh = pblsh instanceof List ? pblsh : [ pblsh ]
+
+    // check elements of publishDir
+    pblsh = pblsh.collect{ elem ->
+      // turn into map if not already so
+      elem = elem instanceof CharSequence ? [ path: elem ] : elem
+
+      // check types and keys
+      assert elem instanceof Map : "Expected publish argument '$elem' to be a String or a Map. Found: class ${elem.getClass()}"
+      assertMapKeys(elem, [ "path", "mode", "overwrite", "pattern", "saveAs", "enabled" ], ["path"], "publishDir")
+
+      // check elements in map
+      assert elem.containsKey("path")
+      assert elem["path"] instanceof CharSequence
+      if (elem.containsKey("mode")) {
+        assert elem["mode"] instanceof CharSequence
+        assert elem["mode"] in [ "symlink", "rellink", "link", "copy", "copyNoFollow", "move" ]
+      }
+      if (elem.containsKey("overwrite")) {
+        assert elem["overwrite"] instanceof Boolean
+      }
+      if (elem.containsKey("pattern")) {
+        assert elem["pattern"] instanceof CharSequence
+      }
+      if (elem.containsKey("saveAs")) {
+        assert elem["saveAs"] instanceof CharSequence //: "saveAs as a Closure is currently not supported. Surround your closure with single quotes to get the desired effect. Example: '\{ foo \}'"
+      }
+      if (elem.containsKey("enabled")) {
+        assert elem["enabled"] instanceof Boolean
+      }
+
+      // return final result
+      elem
+    }
+    // store final directive
+    drctv["publishDir"] = pblsh
+  }
+
+  /* DIRECTIVE queue
+    accepted examples:
+    - "long"
+    - "short,long"
+    - ["short", "long"]
+  */
+  if (drctv.containsKey("queue")) {
+    if (drctv["queue"] instanceof List) {
+      drctv["queue"] = drctv["queue"].join(",")
+    }
+    assert drctv["queue"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE label
+    accepted examples:
+    - "big_mem"
+    - "big_cpu"
+    - ["big_mem", "big_cpu"]
+  */
+  if (drctv.containsKey("label")) {
+    if (drctv["label"] instanceof CharSequence) {
+      drctv["label"] = [ drctv["label"] ]
+    }
+    assert drctv["label"] instanceof List
+    drctv["label"].forEach { label ->
+      assert label instanceof CharSequence
+      // assert label.matches("[a-zA-Z0-9]([a-zA-Z0-9_]*[a-zA-Z0-9])?")
+      // ^ does not allow closures
+    }
+  }
+
+  /* DIRECTIVE scratch
+    accepted examples:
+    - true
+    - "/path/to/scratch"
+    - '$MY_PATH_TO_SCRATCH'
+    - "ram-disk"
+  */
+  if (drctv.containsKey("scratch")) {
+    assert drctv["scratch"] == true || drctv["scratch"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE storeDir
+    accepted examples:
+    - "/path/to/storeDir"
+  */
+  if (drctv.containsKey("storeDir")) {
+    assert drctv["storeDir"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE stageInMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageInMode")) {
+    assert drctv["stageInMode"] instanceof CharSequence
+    assert drctv["stageInMode"] in ["copy", "link", "symlink", "rellink"]
+  }
+
+  /* DIRECTIVE stageOutMode
+    accepted examples:
+    - "copy"
+    - "link"
+  */
+  if (drctv.containsKey("stageOutMode")) {
+    assert drctv["stageOutMode"] instanceof CharSequence
+    assert drctv["stageOutMode"] in ["copy", "move", "rsync"]
+  }
+
+  /* DIRECTIVE tag
+    accepted examples:
+    - "foo"
+    - '$id'
+  */
+  if (drctv.containsKey("tag")) {
+    assert drctv["tag"] instanceof CharSequence
+  }
+
+  /* DIRECTIVE time
+    accepted examples:
+    - "1h"
+    - "2days"
+    - "1day 6hours 3minutes 30seconds"
+  */
+  if (drctv.containsKey("time")) {
+    assert drctv["time"] instanceof CharSequence
+    // todo: validation regex?
+  }
+
+  return drctv
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/processWorkflowArgs.nf'
+def processWorkflowArgs(Map args, Map defaultWfArgs, Map meta) {
+  // override defaults with args
+  def workflowArgs = defaultWfArgs + args
+
+  // check whether 'key' exists
+  assert workflowArgs.containsKey("key") : "Error in module '${meta.config.name}': key is a required argument"
+
+  // if 'key' is a closure, apply it to the original key
+  if (workflowArgs["key"] instanceof Closure) {
+    workflowArgs["key"] = workflowArgs["key"](meta.config.name)
+  }
+  def key = workflowArgs["key"]
+  assert key instanceof CharSequence : "Expected process argument 'key' to be a String. Found: class ${key.getClass()}"
+  assert key ==~ /^[a-zA-Z_]\w*$/ : "Error in module '$key': Expected process argument 'key' to consist of only letters, digits or underscores. Found: ${key}"
+
+  // check for any unexpected keys
+  def expectedKeys = ["key", "directives", "auto", "map", "mapId", "mapData", "mapPassthrough", "filter", "runIf", "fromState", "toState", "args", "renameKeys", "debug"]
+  def unexpectedKeys = workflowArgs.keySet() - expectedKeys
+  assert unexpectedKeys.isEmpty() : "Error in module '$key': unexpected arguments to the '.run()' function: '${unexpectedKeys.join("', '")}'"
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("directives") : "Error in module '$key': directives is a required argument"
+  assert workflowArgs["directives"] instanceof Map : "Error in module '$key': Expected process argument 'directives' to be a Map. Found: class ${workflowArgs['directives'].getClass()}"
+  workflowArgs["directives"] = processDirectives(defaultWfArgs.directives + workflowArgs["directives"])
+
+  // check whether directives exists and apply defaults
+  assert workflowArgs.containsKey("auto") : "Error in module '$key': auto is a required argument"
+  assert workflowArgs["auto"] instanceof Map : "Error in module '$key': Expected process argument 'auto' to be a Map. Found: class ${workflowArgs['auto'].getClass()}"
+  workflowArgs["auto"] = processAuto(defaultWfArgs.auto + workflowArgs["auto"])
+
+  // auto define publish, if so desired
+  if (workflowArgs.auto.publish == true && (workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : [:]).isEmpty()) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.publish is true, params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.publish_dir = \"./output/\""
+    def publishDir = getPublishDir()
+    
+    if (publishDir != null) {
+      workflowArgs.directives.publishDir = [[ 
+        path: publishDir, 
+        saveAs: "{ it.startsWith('.') ? null : it }", // don't publish hidden files, by default
+        mode: "copy"
+      ]]
+    }
+  }
+
+  // auto define transcript, if so desired
+  if (workflowArgs.auto.transcript == true) {
+    // can't assert at this level thanks to the no_publish profile
+    // assert params.containsKey("transcriptsDir") || params.containsKey("transcripts_dir") || params.containsKey("publishDir") || params.containsKey("publish_dir") : 
+    //   "Error in module '${workflowArgs['key']}': if auto.transcript is true, either params.transcripts_dir or params.publish_dir needs to be defined.\n" +
+    //   "  Example: params.transcripts_dir = \"./transcripts/\""
+    def transcriptsDir = 
+      params.containsKey("transcripts_dir") ? params.transcripts_dir : 
+      params.containsKey("transcriptsDir") ? params.transcriptsDir : 
+      params.containsKey("publish_dir") ? params.publish_dir + "/_transcripts" :
+      params.containsKey("publishDir") ? params.publishDir + "/_transcripts" : 
+      null
+    if (transcriptsDir != null) {
+      def timestamp = nextflow.Nextflow.getSession().getWorkflowMetadata().start.format('yyyy-MM-dd_HH-mm-ss')
+      def transcriptsPublishDir = [ 
+        path: "$transcriptsDir/$timestamp/\${task.process.replaceAll(':', '-')}/\${id}/",
+        saveAs: "{ it.startsWith('.') ? it.replaceAll('^.', '') : null }", 
+        mode: "copy"
+      ]
+      def publishDirs = workflowArgs.directives.publishDir != null ? workflowArgs.directives.publishDir : null ? workflowArgs.directives.publishDir : []
+      workflowArgs.directives.publishDir = publishDirs + transcriptsPublishDir
+    }
+  }
+
+  // if this is a stubrun, remove certain directives?
+  if (workflow.stubRun) {
+    workflowArgs.directives.keySet().removeAll(["publishDir", "cpus", "memory", "label"])
+  }
+
+  for (nam in ["map", "mapId", "mapData", "mapPassthrough", "filter", "runIf"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam]) {
+      assert workflowArgs[nam] instanceof Closure : "Error in module '$key': Expected process argument '$nam' to be null or a Closure. Found: class ${workflowArgs[nam].getClass()}"
+    }
+  }
+
+  // TODO: should functions like 'map', 'mapId', 'mapData', 'mapPassthrough' be deprecated as well?
+  for (nam in ["map", "mapData", "mapPassthrough", "renameKeys"]) {
+    if (workflowArgs.containsKey(nam) && workflowArgs[nam] != null) {
+      log.warn "module '$key': workflow argument '$nam' is deprecated and will be removed in Viash 0.9.0. Please use 'fromState' and 'toState' instead."
+    }
+  }
+
+  // check fromState
+  workflowArgs["fromState"] = _processFromState(workflowArgs.get("fromState"), key, meta.config)
+
+  // check toState
+  workflowArgs["toState"] = _processToState(workflowArgs.get("toState"), key, meta.config)
+
+  // return output
+  return workflowArgs
+}
+
+def _processFromState(fromState, key_, config_) {
+  assert fromState == null || fromState instanceof Closure || fromState instanceof Map || fromState instanceof List :
+    "Error in module '$key_': Expected process argument 'fromState' to be null, a Closure, a Map, or a List. Found: class ${fromState.getClass()}"
+  if (fromState == null) {
+    return null
+  }
+  
+  // if fromState is a List, convert to map
+  if (fromState instanceof List) {
+    // check whether fromstate is a list[string]
+    assert fromState.every{it instanceof CharSequence} : "Error in module '$key_': fromState is a List, but not all elements are Strings"
+    fromState = fromState.collectEntries{[it, it]}
+  }
+
+  // if fromState is a map, convert to closure
+  if (fromState instanceof Map) {
+    // check whether fromstate is a map[string, string]
+    assert fromState.values().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all values are Strings"
+    assert fromState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': fromState is a Map, but not all keys are Strings"
+    def fromStateMap = fromState.clone()
+    def requiredInputNames = meta.config.allArguments.findAll{it.required && it.direction == "Input"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    fromState = { it ->
+      def state = it[1]
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def data = fromStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (state.containsKey(origkey)) {
+          [[newkey, state[origkey]]]
+        } else if (!requiredInputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': fromState key '$origkey' not found in current state")
+        }
+      }.collectEntries()
+      data
+    }
+  }
+  
+  return fromState
+}
+
+def _processToState(toState, key_, config_) {
+  if (toState == null) {
+    toState = { tup -> tup[1] }
+  }
+
+  // toState should be a closure, map[string, string], or list[string]
+  assert toState instanceof Closure || toState instanceof Map || toState instanceof List :
+    "Error in module '$key_': Expected process argument 'toState' to be a Closure, a Map, or a List. Found: class ${toState.getClass()}"
+
+  // if toState is a List, convert to map
+  if (toState instanceof List) {
+    // check whether toState is a list[string]
+    assert toState.every{it instanceof CharSequence} : "Error in module '$key_': toState is a List, but not all elements are Strings"
+    toState = toState.collectEntries{[it, it]}
+  }
+
+  // if toState is a map, convert to closure
+  if (toState instanceof Map) {
+    // check whether toState is a map[string, string]
+    assert toState.values().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all values are Strings"
+    assert toState.keySet().every{it instanceof CharSequence} : "Error in module '$key_': toState is a Map, but not all keys are Strings"
+    def toStateMap = toState.clone()
+    def requiredOutputNames = config_.allArguments.findAll{it.required && it.direction == "Output"}.collect{it.plainName}
+    // turn the map into a closure to be used later on
+    toState = { it ->
+      def output = it[1]
+      def state = it[2]
+      assert output instanceof Map : "Error in module '$key_': the output is not a Map"
+      assert state instanceof Map : "Error in module '$key_': the state is not a Map"
+      def extraEntries = toStateMap.collectMany{newkey, origkey ->
+        // check whether newkey corresponds to a required argument
+        if (output.containsKey(origkey)) {
+          [[newkey, output[origkey]]]
+        } else if (!requiredOutputNames.contains(origkey)) {
+          []
+        } else {
+          throw new Exception("Error in module '$key_': toState key '$origkey' not found in current output")
+        }
+      }.collectEntries()
+      state + extraEntries
+    }
+  }
+
+  return toState
+}
+
+// helper file: 'src/main/resources/io/viash/runners/nextflow/workflowFactory/workflowFactory.nf'
+def _debug(workflowArgs, debugKey) {
+  if (workflowArgs.debug) {
+    view { "process '${workflowArgs.key}' $debugKey tuple: $it"  }
+  } else {
+    map { it }
+  }
+}
+
+// depends on: innerWorkflowFactory
+def workflowFactory(Map args, Map defaultWfArgs, Map meta) {
+  def workflowArgs = processWorkflowArgs(args, defaultWfArgs, meta)
+  def key_ = workflowArgs["key"]
+  def multipleArgs = meta.config.allArguments.findAll{ it.multiple }.collect{it.plainName}
+
+  workflow workflowInstance {
+    take: input_
+
+    main:
+    def chModified = input_
+      | checkUniqueIds([:])
+      | _debug(workflowArgs, "input")
+      | map { tuple ->
+        tuple = deepClone(tuple)
+        
+        if (workflowArgs.map) {
+          tuple = workflowArgs.map(tuple)
+        }
+        if (workflowArgs.mapId) {
+          tuple[0] = workflowArgs.mapId(tuple[0])
+        }
+        if (workflowArgs.mapData) {
+          tuple[1] = workflowArgs.mapData(tuple[1])
+        }
+        if (workflowArgs.mapPassthrough) {
+          tuple = tuple.take(2) + workflowArgs.mapPassthrough(tuple.drop(2))
+        }
+
+        // check tuple
+        assert tuple instanceof List : 
+          "Error in module '${key_}': element in channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        assert tuple.size() >= 2 : 
+          "Error in module '${key_}': expected length of tuple in input channel to be two or greater.\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: tuple.size() == ${tuple.size()}"
+        
+        // check id field
+        if (tuple[0] instanceof GString) {
+          tuple[0] = tuple[0].toString()
+        }
+        assert tuple[0] instanceof CharSequence : 
+          "Error in module '${key_}': first element of tuple in channel should be a String\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Found: ${tuple[0]}"
+        
+        // match file to input file
+        if (workflowArgs.auto.simplifyInput && (tuple[1] instanceof Path || tuple[1] instanceof List)) {
+          def inputFiles = meta.config.allArguments
+            .findAll { it.type == "file" && it.direction == "input" }
+          
+          assert inputFiles.size() == 1 : 
+              "Error in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Anonymous file inputs are only allowed when the process has exactly one file input.\n" +
+              "  Expected: inputFiles.size() == 1. Found: inputFiles.size() is ${inputFiles.size()}"
+
+          tuple[1] = [[ inputFiles[0].plainName, tuple[1] ]].collectEntries()
+        }
+
+        // check data field
+        assert tuple[1] instanceof Map : 
+          "Error in module '${key_}' id '${tuple[0]}': second element of tuple in channel should be a Map\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+        // rename keys of data field in tuple
+        if (workflowArgs.renameKeys) {
+          assert workflowArgs.renameKeys instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class: Map. Found: renameKeys.getClass() is ${workflowArgs.renameKeys.getClass()}"
+          assert tuple[1] instanceof Map : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Expected class: Map. Found: tuple[1].getClass() is ${tuple[1].getClass()}"
+
+          // TODO: allow renameKeys to be a function?
+          workflowArgs.renameKeys.each { newKey, oldKey ->
+            assert newKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of newKey: String. Found: newKey.getClass() is ${newKey.getClass()}"
+            assert oldKey instanceof CharSequence : 
+              "Error renaming data keys in module '${key_}' id '${tuple[0]}'.\n" +
+              "  Example: renameKeys: ['new_key': 'old_key'].\n" +
+              "  Expected class of oldKey: String. Found: oldKey.getClass() is ${oldKey.getClass()}"
+            assert tuple[1].containsKey(oldKey) : 
+              "Error renaming data keys in module '${key}' id '${tuple[0]}'.\n" +
+              "  Key '$oldKey' is missing in the data map. tuple[1].keySet() is '${tuple[1].keySet()}'"
+            tuple[1].put(newKey, tuple[1][oldKey])
+          }
+          tuple[1].keySet().removeAll(workflowArgs.renameKeys.collect{ newKey, oldKey -> oldKey })
+        }
+        tuple
+      }
+
+
+    def chRun = null
+    def chPassthrough = null
+    if (workflowArgs.runIf) {
+      def runIfBranch = chModified.branch{ tup ->
+        run: workflowArgs.runIf(tup[0], tup[1])
+        passthrough: true
+      }
+      chRun = runIfBranch.run
+      chPassthrough = runIfBranch.passthrough
+    } else {
+      chRun = chModified
+      chPassthrough = Channel.empty()
+    }
+
+    def chRunFiltered = workflowArgs.filter ?
+      chRun | filter{workflowArgs.filter(it)} :
+      chRun
+
+    def chArgs = workflowArgs.fromState ? 
+      chRunFiltered | map{
+        def new_data = workflowArgs.fromState(it.take(2))
+        [it[0], new_data]
+      } :
+      chRunFiltered | map {tup -> tup.take(2)}
+
+    // fill in defaults
+    def chArgsWithDefaults = chArgs
+      | map { tuple ->
+        def id_ = tuple[0]
+        def data_ = tuple[1]
+
+        // TODO: could move fromState to here
+
+        // fetch default params from functionality
+        def defaultArgs = meta.config.allArguments
+          .findAll { it.containsKey("default") }
+          .collectEntries { [ it.plainName, it.default ] }
+
+        // fetch overrides in params
+        def paramArgs = meta.config.allArguments
+          .findAll { par ->
+            def argKey = key_ + "__" + par.plainName
+            params.containsKey(argKey)
+          }
+          .collectEntries { [ it.plainName, params[key_ + "__" + it.plainName] ] }
+        
+        // fetch overrides in data
+        def dataArgs = meta.config.allArguments
+          .findAll { data_.containsKey(it.plainName) }
+          .collectEntries { [ it.plainName, data_[it.plainName] ] }
+        
+        // combine params
+        def combinedArgs = defaultArgs + paramArgs + workflowArgs.args + dataArgs
+
+        // remove arguments with explicit null values
+        combinedArgs
+          .removeAll{_, val -> val == null || val == "viash_no_value" || val == "force_null"}
+
+        combinedArgs = _processInputValues(combinedArgs, meta.config, id_, key_)
+
+        [id_, combinedArgs] + tuple.drop(2)
+      }
+
+    // TODO: move some of the _meta.join_id wrangling to the safeJoin() function.
+    def chInitialOutputMulti = chArgsWithDefaults
+      | _debug(workflowArgs, "processed")
+      // run workflow
+      | innerWorkflowFactory(workflowArgs)
+    def chInitialOutputList = chInitialOutputMulti instanceof List ? chInitialOutputMulti : [chInitialOutputMulti]
+    assert chInitialOutputList.size() > 0: "should have emitted at least one output channel"
+    // Add a channel ID to the events, which designates the channel the event was emitted from as a running number
+    // This number is used to sort the events later when the events are gathered from across the channels.
+    def chInitialOutputListWithIndexedEvents = chInitialOutputList.withIndex().collect{channel, channelIndex ->
+      def newChannel = channel
+        | map {tuple ->
+          assert tuple instanceof List : 
+          "Error in module '${key_}': element in output channel should be a tuple [id, data, ...otherargs...]\n" +
+          "  Example: [\"id\", [input: file('foo.txt'), arg: 10]].\n" +
+          "  Expected class: List. Found: tuple.getClass() is ${tuple.getClass()}"
+        
+          def newEvent = [channelIndex] + tuple
+          return newEvent
+        }
+      return newChannel
+    }
+    // Put the events into 1 channel, cover case where there is only one channel is emitted
+    def chInitialOutput = chInitialOutputList.size() > 1 ? \
+      chInitialOutputListWithIndexedEvents[0].mix(*chInitialOutputListWithIndexedEvents.tail()) : \
+      chInitialOutputListWithIndexedEvents[0]
+    def chInitialOutputProcessed = chInitialOutput
+      | map { tuple  ->
+        def channelId = tuple[0]
+        def id_ = tuple[1]
+        def output_ = tuple[2]
+
+        // see if output map contains metadata
+        def meta_ =
+          output_ instanceof Map && output_.containsKey("_meta") ? 
+          output_["_meta"] :
+          [:]
+        def join_id = meta_.join_id ?: id_
+        
+        // remove metadata
+        output_ = output_.findAll{k, v -> k != "_meta"}
+
+        // check value types
+        output_ = _checkValidOutputArgument(output_, meta.config, id_, key_)
+
+        [join_id, channelId, id_, output_]
+      }
+      // | view{"chInitialOutput: ${it.take(3)}"}
+
+    // join the output [prev_id, channel_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chPublishWithPreviousState = safeJoin(chInitialOutputProcessed, chRunFiltered, key_)
+      // input tuple format: [join_id, channel_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, channel_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(2).take(3))
+        tup.take(3) + [new_state] + tup.drop(5)
+      }
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishFiles = chPublishWithPreviousState
+        // input tuple format: [join_id, channel_id, id, new_state, ...]
+        // output tuple format: [join_id, channel_id, id, new_state]
+        | map{ tup ->
+          tup.take(4)
+        }
+
+      safeJoin(chPublishFiles, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, channel_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(2).take(3)
+        }
+        | publishFilesByConfig(key: key_, config: meta.config)
+    }
+    // Join the state from the events that were emitted from different channels
+    def chJoined = chInitialOutputProcessed
+      | map {tuple ->
+        def join_id = tuple[0]
+        def channel_id = tuple[1]
+        def id = tuple[2]
+        def other = tuple.drop(3)
+        // Below, groupTuple is used to join the events. To make sure resuming a workflow
+        // keeps working, the output state must be deterministic. This means the state needs to be
+        // sorted with groupTuple's has a 'sort' argument. This argument can be set to 'hash',
+        // but hashing the state when it is large can be problematic in terms of performance.
+        // Therefore, a custom comparator function is provided. We add the channel ID to the 
+        // states so that we can use the channel ID to sort the items. 
+        def stateWithChannelID = [[channel_id] * other.size(), other].transpose()
+        // A comparator that is provided to groupTuple's 'sort' argument is applied
+        // to all elements of the event tuple (that is not the 'id'). The comparator
+        // closure that is used below expects the input to be List. So the join_id and
+        // channel_id must also be wrapped in a list. 
+        [[join_id], [channel_id], id] + stateWithChannelID
+      }
+      | groupTuple(by: 2, sort: {a, b -> a[0] <=> b[0]}, size: chInitialOutputList.size(), remainder: true)
+      | map {join_ids, _, id, statesWithChannelID ->
+        // Remove the channel IDs from the states
+        def states = statesWithChannelID.collect{it[1]}
+        def newJoinId = join_ids.flatten().unique{a, b -> a <=> b}
+        assert newJoinId.size() == 1: "Multiple events were emitted for '$id'."
+        def newJoinIdUnique = newJoinId[0]
+        
+        // Merge the states from the different channels
+        def newState = states.inject([:]){ old_state, state_to_add ->
+          return old_state + state_to_add.collectEntries{k, v -> 
+            if (!multipleArgs.contains(k)) {
+              // if the key is not a multiple argument, we expect only one value
+              if (old_state.containsKey(k)) {
+                assert old_state[k] == v : "ID $id: multiple entries for argument $k were emitted."
+              }
+              [k, v]
+            } else {
+              // if the key is a multiple argument, append the different values into one list
+              def prevValue = old_state.getOrDefault(k, [])
+              def prevValueAsList = prevValue instanceof List ? prevValue : [prevValue]
+              [k, prevValueAsList + v]
+            }
+          }
+        }
+
+        _checkAllRequiredOuputsPresent(newState, meta.config, id, key_)
+
+        // simplify output if need be
+        if (workflowArgs.auto.simplifyOutput && newState.size() == 1) {
+          newState = newState.values()[0]
+        }
+
+        return [newJoinIdUnique, id, newState]
+      }
+    
+    // join the output [prev_id, new_id, output] with the previous state [prev_id, state, ...]
+    def chNewState = safeJoin(chJoined, chRunFiltered, key_)
+      // input tuple format: [join_id, id, output, prev_state, ...]
+      // output tuple format: [join_id, id, new_state, ...]
+      | map{ tup ->
+        def new_state = workflowArgs.toState(tup.drop(1).take(3))
+        tup.take(2) + [new_state] + tup.drop(4)
+      }
+
+    if (workflowArgs.auto.publish == "state") {
+      def chPublishStates = chNewState
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [join_id, id, new_state]
+        | map{ tup ->
+          tup.take(3)
+        }
+
+      safeJoin(chPublishStates, chArgsWithDefaults, key_)
+        // input tuple format: [join_id, id, new_state, orig_state, ...]
+        // output tuple format: [id, new_state, orig_state]
+        | map { tup ->
+          tup.drop(1).take(3)
+        }
+        | publishStatesByConfig(key: key_, config: meta.config)
+    }
+    chReturn = chNewState
+      | map { tup ->
+        // input tuple format: [join_id, id, new_state, ...]
+        // output tuple format: [id, new_state, ...]
+        tup.drop(1)
+      }
+      | _debug(workflowArgs, "output")
+      | concat(chPassthrough)
+
+    emit: chReturn
+  }
+
+  def wf = workflowInstance.cloneWithName(key_)
+
+  // add factory function
+  wf.metaClass.run = { runArgs ->
+    workflowFactory(runArgs, workflowArgs, meta)
+  }
+  // add config to module for later introspection
+  wf.metaClass.config = meta.config
+
+  return wf
+}
+
+nextflow.enable.dsl=2
+
+// START COMPONENT-SPECIFIC CODE
+
+// create meta object
+meta = [
+  "resources_dir": moduleDir.toRealPath().normalize(),
+  "config": processConfig(readJsonBlob('''{
+  "name" : "process_integrate_annotate",
+  "namespace" : "single_cell",
+  "version" : "set-public-scope",
+  "authors" : [
+    {
+      "name" : "Dorien Roosen",
+      "roles" : [
+        "author",
+        "maintainer"
+      ],
+      "info" : {
+        "role" : "Core Team Member",
+        "links" : {
+          "email" : "dorien@data-intuitive.com",
+          "github" : "dorien-er",
+          "linkedin" : "dorien-roosen"
+        },
+        "organizations" : [
+          {
+            "name" : "Data Intuitive",
+            "href" : "https://www.data-intuitive.com",
+            "role" : "Data Scientist"
+          }
+        ]
+      }
+    },
+    {
+      "name" : "Weiwei Schultz",
+      "roles" : [
+        "contributor"
+      ],
+      "info" : {
+        "role" : "Contributor",
+        "organizations" : [
+          {
+            "name" : "Janssen R&D US",
+            "role" : "Associate Director Data Sciences"
+          }
+        ]
+      }
+    }
+  ],
+  "argument_groups" : [
+    {
+      "name" : "Input (query) data arguments",
+      "description" : "The input query dataset(s) to be annotated",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--id",
+          "description" : "ID of the sample.",
+          "example" : [
+            "foo"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "file",
+          "name" : "--input",
+          "description" : "Input query dataset(s) to be annotated",
+          "example" : [
+            "input.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--modality",
+          "description" : "Modality to be processed. Should match the modality in the --reference dataset, if provided.",
+          "default" : [
+            "rna"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_layer",
+          "description" : "The layer in the input data containing the raw counts, if .X is not to be used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--input_var_gene_names",
+          "description" : "The name of the adata var column containing gene names; when no gene_name_layer is provided, the var index will be used.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--input_reference_gene_overlap",
+          "description" : "The minimum number of genes present in both the reference and query datasets.\n",
+          "default" : [
+            100
+          ],
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Reference data arguments",
+      "description" : "Dataset to be used as a reference for label transfer and to train annotation algorithms on.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--reference",
+          "description" : "The reference dataset in .h5mu format to be used as a reference mapper and to train annotation algorithms on.\n",
+          "example" : [
+            "reference.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_layer_raw_counts",
+          "description" : "The layer in the reference dataset containing the raw counts, if .X is not to be used.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_layer_lognormalized_counts",
+          "description" : "The layer in the reference dataset containing the log-normalized counts, if .X is not to be used.",
+          "default" : [
+            "log_normalized"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_var_gene_names",
+          "description" : "The name of the adata .var column containing gene names if the .var index is not to be used.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_obs_batch",
+          "description" : "The .obs column of the reference dataset containing the batch information.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_obs_label",
+          "description" : "The `.obs` key of the target labels to tranfer.",
+          "example" : [
+            "cell_type"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_obs_label_unlabeled_category",
+          "description" : "Value in the --reference_obs_label field that indicates unlabeled observations",
+          "default" : [
+            "Unkown"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--reference_var_input",
+          "description" : ".var column containing highly variable genes. By default, do not subset genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Methods",
+      "description" : "The available annotation and integration methods to integrate and/or annotate the query dataset(s) with.",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--integration_methods",
+          "description" : "Integration methods to be executed.",
+          "example" : [
+            "harmony;scvi"
+          ],
+          "required" : false,
+          "choices" : [
+            "harmony",
+            "scvi"
+          ],
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--annotation_methods",
+          "description" : "Annotation methods to be executed.",
+          "example" : [
+            "celltypist;scanvi_scarches"
+          ],
+          "required" : false,
+          "choices" : [
+            "celltypist",
+            "scanvi_scarches"
+          ],
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Pre-processing options: RNA filtering",
+      "description" : "Pre-processing options for filtering RNA data",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--rna_min_counts",
+          "description" : "Minimum number of counts captured per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--rna_max_counts",
+          "description" : "Maximum number of counts captured per cell.",
+          "example" : [
+            5000000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--rna_min_genes_per_cell",
+          "description" : "Minimum of non-zero values per cell.",
+          "example" : [
+            200
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--rna_max_genes_per_cell",
+          "description" : "Maximum of non-zero values per cell.",
+          "example" : [
+            1500000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--rna_min_cells_per_gene",
+          "description" : "Minimum of non-zero values per gene.",
+          "example" : [
+            3
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--rna_min_fraction_mito",
+          "description" : "Minimum fraction of UMIs that are mitochondrial.",
+          "example" : [
+            0.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--rna_max_fraction_mito",
+          "description" : "Maximum fraction of UMIs that are mitochondrial.",
+          "example" : [
+            0.2
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Pre-processing options: Highly variable features detection",
+      "description" : "Pre-processing options for detecting highly variable features",
+      "arguments" : [
+        {
+          "type" : "integer",
+          "name" : "--n_hvg",
+          "description" : "Number of highly-variable features to keep. \nOnly relevant if HVG need to be calculated across query and reference datasets (e.g. for --annotation_methods scvi_knn and harmony_knn). \nFor reference mapping-based methods, the HVG's specified in --reference_var_input will be used.\n",
+          "default" : [
+            2000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Pre-processing options: Mitochondrial & Ribosomal Gene Detection",
+      "description" : "Pre-processing options for detecting mitochondrial genes",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_name_mitochondrial_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--var_name_ribosomal_genes",
+          "description" : "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_mitochondrial_fraction",
+          "description" : "When specified, write the fraction of counts originating from mitochondrial genes \n(based on --mitochondrial_gene_regex) to an .obs column with the specified name.\nRequires --var_name_mitochondrial_genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--obs_name_ribosomal_fraction",
+          "description" : "When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified name.\nRequires --var_name_ribosomal_genes.\n",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--mitochondrial_gene_regex",
+          "description" : "Regex string that identifies mitochondrial genes from --var_gene_names.\nBy default will detect human and mouse mitochondrial genes from a gene symbol.\n",
+          "default" : [
+            "^[mM][tT]-"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--ribosomal_gene_regex",
+          "description" : "Regex string that identifies ribosomal genes from --var_gene_names.\nBy default will detect human and mouse ribosomal genes from a gene symbol.\n",
+          "default" : [
+            "^[Mm]?[Rr][Pp][LlSs]"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Pre-processing options: QC metrics calculation options",
+      "description" : "Pre-processing options for calculating QC metrics",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--var_qc_metrics",
+          "description" : "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes. Defaults to the combined values specified for\n--var_name_mitochondrial_genes and --highly_variable_features_var_output.\n",
+          "example" : [
+            "ercc,highly_variable"
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ","
+        }
+      ]
+    },
+    {
+      "name" : "Harmony integration options",
+      "description" : "Specifications for harmony integration.",
+      "arguments" : [
+        {
+          "type" : "double",
+          "name" : "--harmony_theta",
+          "description" : "Diversity clustering penalty parameter. Specify for each variable in group.by.vars. \ntheta=0 does not encourage any diversity. Larger values of theta\nresult in more diverse clusters.\\"\n",
+          "example" : [
+            0.0,
+            1.0,
+            2.0
+          ],
+          "default" : [
+            2.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--harmony_obs_covariates",
+          "description" : "The .obs field(s) that define the covariate(s) to regress out.",
+          "example" : [
+            "batch",
+            "sample"
+          ],
+          "default" : [
+            "sample_id"
+          ],
+          "required" : true,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "scVI, scANVI and scArches training options",
+      "description" : "Training arguments for scVI, scANVI and scArches. Relevant for --annotation_methods 'scvi_knn' and 'scanvi_scarches'.",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--early_stopping",
+          "description" : "Whether to perform early stopping with respect to the validation set.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "string",
+          "name" : "--early_stopping_monitor",
+          "description" : "Metric logged during validation set epoch.",
+          "default" : [
+            "elbo_validation"
+          ],
+          "required" : false,
+          "choices" : [
+            "elbo_validation",
+            "reconstruction_loss_validation",
+            "kl_local_validation"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--early_stopping_patience",
+          "description" : "Number of validation epochs with no improvement after which training will be stopped.",
+          "default" : [
+            45
+          ],
+          "required" : false,
+          "min" : 1,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--early_stopping_min_delta",
+          "description" : "Minimum change in the monitored quantity to qualify as an improvement, i.e. an absolute change of less than min_delta, will count as no improvement.",
+          "default" : [
+            0.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--max_epochs",
+          "description" : "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest.",
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--reduce_lr_on_plateau",
+          "description" : "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus.",
+          "default" : [
+            true
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--lr_factor",
+          "description" : "Factor to reduce learning rate.",
+          "default" : [
+            0.6
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--lr_patience",
+          "description" : "Number of epochs with no improvement after which learning rate will be reduced.",
+          "default" : [
+            30.0
+          ],
+          "required" : false,
+          "min" : 0.0,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "CellTypist reference model",
+      "description" : "The CellTypist reference model to use for annotation. If not provided, the reference dataset will be used for model training.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--celltypist_model",
+          "description" : "Pretrained model in pkl format. If not provided, the model will be trained on the reference data and --reference should be provided.",
+          "example" : [
+            "pretrained_model.pkl"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "CellTypist annotation options",
+      "description" : "Specifications for CellTypist annotation.",
+      "arguments" : [
+        {
+          "type" : "boolean",
+          "name" : "--celltypist_feature_selection",
+          "description" : "Whether to perform feature selection.",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean",
+          "name" : "--celltypist_majority_voting",
+          "description" : "Whether to refine the predicted labels by running the majority voting classifier after over-clustering.",
+          "default" : [
+            false
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "double",
+          "name" : "--celltypist_C",
+          "description" : "Inverse of regularization strength in logistic regression.",
+          "default" : [
+            1.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--celltypist_max_iter",
+          "description" : "Maximum number of iterations before reaching the minimum of the cost function.",
+          "default" : [
+            1000
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "boolean_true",
+          "name" : "--celltypist_use_SGD",
+          "description" : "Whether to use the stochastic gradient descent algorithm.",
+          "direction" : "input"
+        },
+        {
+          "type" : "double",
+          "name" : "--celltypist_min_prop",
+          "description" : "\\"For the dominant cell type within a subcluster, the minimum proportion of cells required to \nsupport naming of the subcluster by this cell type. Ignored if majority_voting is set to False. \nSubcluster that fails to pass this proportion threshold will be assigned 'Heterogeneous'.\\"\n",
+          "default" : [
+            0.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Clustering options",
+      "description" : "Arguments for Leiden clustering. Only relevant for --annotation_methods `scvi_knn`, `scanvi_scarches` and `harmony_knn`.",
+      "arguments" : [
+        {
+          "type" : "double",
+          "name" : "--leiden_resolution",
+          "description" : "Control the coarseness of the clustering. Higher values lead to more clusters.",
+          "default" : [
+            1.0
+          ],
+          "required" : false,
+          "direction" : "input",
+          "multiple" : true,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Neighbor classifier arguments",
+      "description" : "Arguments related to calculating the n nearest neighbors. Only relevant for --annotation_methods `scvi_knn`, `scanvi_scarches` and `harmony_knn`.",
+      "arguments" : [
+        {
+          "type" : "string",
+          "name" : "--knn_weights",
+          "description" : "Weight function used in prediction. Possible values are:\n`uniform` (all points in each neighborhood are weighted equally) or \n`distance` (weight points by the inverse of their distance)\n",
+          "default" : [
+            "uniform"
+          ],
+          "required" : false,
+          "choices" : [
+            "uniform",
+            "distance"
+          ],
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        },
+        {
+          "type" : "integer",
+          "name" : "--knn_n_neighbors",
+          "description" : "The number of neighbors to use in k-neighbor graph structure used for fast approximate nearest neighbor search with PyNNDescent. \nLarger values will result in more accurate search results at the cost of computation time.\n",
+          "default" : [
+            15
+          ],
+          "required" : false,
+          "min" : 5,
+          "direction" : "input",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    },
+    {
+      "name" : "Outputs",
+      "description" : "The output file to write the annotated dataset to.",
+      "arguments" : [
+        {
+          "type" : "file",
+          "name" : "--output",
+          "description" : "The output file.\n",
+          "example" : [
+            "output.h5mu"
+          ],
+          "must_exist" : true,
+          "create_parent" : true,
+          "required" : true,
+          "direction" : "output",
+          "multiple" : false,
+          "multiple_sep" : ";"
+        }
+      ]
+    }
+  ],
+  "resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "main.nf",
+      "is_executable" : true,
+      "entrypoint" : "run_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/src/configs/labels.config",
+      "dest" : "nextflow_labels.config"
+    }
+  ],
+  "description" : "A pipeline to process, integrate and annotate single cell (multi-)omics data.\nAvailable integration methods:\n- Harmony\n- scVI\nAvailable annotation methods:\n- CellTypist\n- scANVI (with scArches)\n",
+  "test_resources" : [
+    {
+      "type" : "nextflow_script",
+      "path" : "test.nf",
+      "is_executable" : true,
+      "entrypoint" : "test_wf"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/pbmc_1k_protein_v3/pbmc_1k_protein_v3_mms.h5mu"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/annotation_test_data/TS_Blood_filtered.h5mu"
+    },
+    {
+      "type" : "file",
+      "path" : "/resources_test/annotation_test_data/celltypist_model_Immune_All_Low.pkl"
+    }
+  ],
+  "status" : "enabled",
+  "scope" : {
+    "image" : "public",
+    "target" : "public"
+  },
+  "requirements" : {
+    "commands" : [
+      "ps"
+    ]
+  },
+  "dependencies" : [
+    {
+      "name" : "workflows/multiomics/process_samples",
+      "alias" : "process_samples_workflow",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.4"
+      }
+    },
+    {
+      "name" : "annotate/celltypist",
+      "alias" : "celltypist_annotation",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.4"
+      }
+    },
+    {
+      "name" : "workflows/annotation/scanvi_scarches",
+      "alias" : "scanvi_scarches_annotation",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.4"
+      }
+    },
+    {
+      "name" : "workflows/integration/harmony_leiden",
+      "alias" : "harmony_integration",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.4"
+      }
+    },
+    {
+      "name" : "workflows/integration/scvi_leiden",
+      "alias" : "scvi_integration",
+      "repository" : {
+        "type" : "vsh",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.4"
+      }
+    }
+  ],
+  "repositories" : [
+    {
+      "type" : "vsh",
+      "name" : "openpipeline",
+      "repo" : "openpipeline",
+      "tag" : "v4.0.4"
+    },
+    {
+      "type" : "vsh",
+      "name" : "openpipeline_qc",
+      "repo" : "openpipeline_qc",
+      "tag" : "v0.2.2"
+    },
+    {
+      "type" : "vsh",
+      "name" : "biobox",
+      "repo" : "biobox",
+      "tag" : "v0.4.2"
+    }
+  ],
+  "links" : {
+    "repository" : "https://github.com/openpipelines-bio/openpipeline_composed",
+    "docker_registry" : "ghcr.io"
+  },
+  "runners" : [
+    {
+      "type" : "nextflow",
+      "id" : "nextflow",
+      "directives" : {
+        "tag" : "$id"
+      },
+      "auto" : {
+        "simplifyInput" : true,
+        "simplifyOutput" : false,
+        "transcript" : false,
+        "publish" : false
+      },
+      "config" : {
+        "labels" : {
+          "mem1gb" : "memory = 1000000000.B",
+          "mem2gb" : "memory = 2000000000.B",
+          "mem5gb" : "memory = 5000000000.B",
+          "mem10gb" : "memory = 10000000000.B",
+          "mem20gb" : "memory = 20000000000.B",
+          "mem50gb" : "memory = 50000000000.B",
+          "mem100gb" : "memory = 100000000000.B",
+          "mem200gb" : "memory = 200000000000.B",
+          "mem500gb" : "memory = 500000000000.B",
+          "mem1tb" : "memory = 1000000000000.B",
+          "mem2tb" : "memory = 2000000000000.B",
+          "mem5tb" : "memory = 5000000000000.B",
+          "mem10tb" : "memory = 10000000000000.B",
+          "mem20tb" : "memory = 20000000000000.B",
+          "mem50tb" : "memory = 50000000000000.B",
+          "mem100tb" : "memory = 100000000000000.B",
+          "mem200tb" : "memory = 200000000000000.B",
+          "mem500tb" : "memory = 500000000000000.B",
+          "mem1gib" : "memory = 1073741824.B",
+          "mem2gib" : "memory = 2147483648.B",
+          "mem4gib" : "memory = 4294967296.B",
+          "mem8gib" : "memory = 8589934592.B",
+          "mem16gib" : "memory = 17179869184.B",
+          "mem32gib" : "memory = 34359738368.B",
+          "mem64gib" : "memory = 68719476736.B",
+          "mem128gib" : "memory = 137438953472.B",
+          "mem256gib" : "memory = 274877906944.B",
+          "mem512gib" : "memory = 549755813888.B",
+          "mem1tib" : "memory = 1099511627776.B",
+          "mem2tib" : "memory = 2199023255552.B",
+          "mem4tib" : "memory = 4398046511104.B",
+          "mem8tib" : "memory = 8796093022208.B",
+          "mem16tib" : "memory = 17592186044416.B",
+          "mem32tib" : "memory = 35184372088832.B",
+          "mem64tib" : "memory = 70368744177664.B",
+          "mem128tib" : "memory = 140737488355328.B",
+          "mem256tib" : "memory = 281474976710656.B",
+          "mem512tib" : "memory = 562949953421312.B",
+          "cpu1" : "cpus = 1",
+          "cpu2" : "cpus = 2",
+          "cpu5" : "cpus = 5",
+          "cpu10" : "cpus = 10",
+          "cpu20" : "cpus = 20",
+          "cpu50" : "cpus = 50",
+          "cpu100" : "cpus = 100",
+          "cpu200" : "cpus = 200",
+          "cpu500" : "cpus = 500",
+          "cpu1000" : "cpus = 1000"
+        },
+        "script" : [
+          "includeConfig(\\"nextflow_labels.config\\")"
+        ]
+      },
+      "debug" : false,
+      "container" : "docker"
+    }
+  ],
+  "engines" : [
+    {
+      "type" : "native",
+      "id" : "native"
+    }
+  ],
+  "build_info" : {
+    "config" : "/workdir/root/repo/src/single_cell/process_integrate_annotate/config.vsh.yaml",
+    "runner" : "nextflow",
+    "engine" : "native",
+    "output" : "/workdir/root/repo/target/nextflow/single_cell/process_integrate_annotate",
+    "viash_version" : "0.9.7",
+    "git_commit" : "50682718d61f38677e46dde857fb948846ec6cf6",
+    "git_remote" : "https://github.com/openpipelines-bio/openpipeline_composed"
+  },
+  "package_config" : {
+    "name" : "openpipeline_composed",
+    "version" : "set-public-scope",
+    "info" : {
+      "test_resources" : [
+        {
+          "type" : "s3",
+          "path" : "s3://openpipelines-bio/openpipeline_incubator/resources_test",
+          "dest" : "resources_test"
+        }
+      ]
+    },
+    "repositories" : [
+      {
+        "type" : "vsh",
+        "name" : "openpipeline",
+        "repo" : "openpipeline",
+        "tag" : "v4.0.4"
+      },
+      {
+        "type" : "vsh",
+        "name" : "openpipeline_qc",
+        "repo" : "openpipeline_qc",
+        "tag" : "v0.2.2"
+      },
+      {
+        "type" : "vsh",
+        "name" : "biobox",
+        "repo" : "biobox",
+        "tag" : "v0.4.2"
+      }
+    ],
+    "viash_version" : "0.9.7",
+    "source" : "/workdir/root/repo/src",
+    "target" : "/workdir/root/repo/target",
+    "config_mods" : [
+      ".requirements.commands := ['ps']\n.runners[.type == 'nextflow'].directives.tag := '$id'\n.resources += {path: '/src/configs/labels.config', dest: 'nextflow_labels.config'}\n.runners[.type == 'nextflow'].config.script := 'includeConfig(\\"nextflow_labels.config\\")'\n",
+      ".engines += { type: \\"native\\" }",
+      ".engines[.type == 'docker'].target_registry := 'images.viash-hub.com'",
+      ".engines[.type == 'docker'].target_tag := 'set-public-scope'"
+    ],
+    "organization" : "vsh",
+    "links" : {
+      "repository" : "https://github.com/openpipelines-bio/openpipeline_composed",
+      "docker_registry" : "ghcr.io"
+    }
+  }
+}'''))
+]
+
+// resolve dependencies dependencies (if any)
+meta["root_dir"] = getRootDir()
+include { process_samples as process_samples_workflow_viashalias } from "${meta.root_dir}/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/multiomics/process_samples/main.nf"
+process_samples_workflow = process_samples_workflow_viashalias.run(key: "process_samples_workflow")
+include { celltypist as celltypist_annotation_viashalias } from "${meta.root_dir}/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/annotate/celltypist/main.nf"
+celltypist_annotation = celltypist_annotation_viashalias.run(key: "celltypist_annotation")
+include { scanvi_scarches as scanvi_scarches_annotation_viashalias } from "${meta.root_dir}/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/annotation/scanvi_scarches/main.nf"
+scanvi_scarches_annotation = scanvi_scarches_annotation_viashalias.run(key: "scanvi_scarches_annotation")
+include { harmony_leiden as harmony_integration_viashalias } from "${meta.root_dir}/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/harmony_leiden/main.nf"
+harmony_integration = harmony_integration_viashalias.run(key: "harmony_integration")
+include { scvi_leiden as scvi_integration_viashalias } from "${meta.root_dir}/dependencies/vsh/vsh/openpipeline/v4.0.4/nextflow/workflows/integration/scvi_leiden/main.nf"
+scvi_integration = scvi_integration_viashalias.run(key: "scvi_integration")
+
+// inner workflow
+// user-provided Nextflow code
+workflow run_wf {
+  take:
+    input_ch
+
+  main:
+    output_ch = input_ch
+    | map { id, state ->
+      def new_state = state + [ "query_processed": state.output, "_meta": ["join_id": id] ]
+      [id, new_state]
+    }
+    // Make sure parameters are filled out correctly
+    | map { id, state ->
+      def new_state = [:]
+      // Check that at least one of annotation_methods or integration_methods is not empty
+      if (!state.annotation_methods  && !state.integration_methods) {
+        throw new RuntimeException("At least one of --annotation_methods or --integration_methods must be provided")
+      }
+      // Check CellTypist arguments
+      if (state.annotation_methods && state.annotation_methods.contains("celltypist") && 
+        (!state.celltypist_model && !state.reference)) {
+        throw new RuntimeException("Celltypist was selected as an annotation method. Either --celltypist_model or --reference must be provided.")
+      }
+      if (state.annotation_methods && state.annotation_methods.contains("celltypist") && state.celltypist_model && state.reference )  {
+        System.err.println(
+          "Warning: --celltypist_model is set and a --reference was provided. \
+          The pre-trained Celltypist model will be used for annotation, the reference will be ignored."
+        )
+      }
+
+      [id, state + new_state]
+    }
+    | process_samples_workflow.run(
+      fromState: [
+        "input": "input", 
+        "id": "id",
+        "rna_layer": "input_layer",
+        "rna_min_counts": "rna_min_counts",
+        "rna_max_counts": "rna_max_counts",
+        "rna_min_genes_per_cell": "rna_min_genes_per_cell",
+        "rna_max_genes_per_cell": "rna_max_genes_per_cell",
+        "rna_min_cells_per_gene": "rna_min_cells_per_gene",
+        "rna_min_fraction_mito": "rna_min_fraction_mito",
+        "rna_max_fraction_mito": "rna_max_fraction_mito",
+        "rna_min_fraction_ribo": "rna_min_fraction_ribo",
+        "rna_max_fraction_ribo": "rna_max_fraction_ribo",
+        "var_name_mitochondrial_genes": "var_name_mitochondrial_genes",
+        "var_name_ribosomal_genes": "var_name_ribosomal_genes",
+        "var_gene_names": "input_var_gene_names",
+        "mitochondrial_gene_regex": "mitochondrial_gene_regex",
+        "ribosomal_gene_regex": "ribosomal_gene_regex",
+        "var_qc_metrics": "var_qc_metrics"
+      ],
+      args: [
+        "pca_overwrite": "true",
+        "add_id_obs_output": "sample_id",
+        "highly_variable_features_var_output": "filter_with_hvg_query"
+      ],
+      toState: ["query_processed": "output"], 
+    )
+    // Integration methods
+    | harmony_integration.run(
+      runIf: { id, state -> 
+        state.integration_methods && state.integration_methods.contains("harmony") 
+      },
+      fromState: [ 
+        "id": "id",
+        "input": "query_processed",
+        "modality": "modality",
+        "theta": "harmony_theta",
+        "leiden_resolution": "leiden_resolution",
+        "obs_covariates": "harmony_obs_covariates"
+      ],
+      args: [
+        "layer": "log_normalized",
+        "embedding": "X_pca",
+        "obsm_integrated": "X_harmony_integrated",
+        "uns_neighbors": "harmony_integration_neighbors",
+        "obsp_neighbor_distances": "harmony_integration_neighbor_distances",
+        "obsp_neighbor_connectivities": "harmony_integration_neighbor_connectivities",
+        "obs_cluster": "harmony_integration_leiden",
+        "obsm_umap": "X_harmony_umap"
+      ],
+      toState: [ "query_processed": "output" ]
+    )
+
+    | scvi_integration.run(
+      runIf: { id, state -> 
+        state.integration_methods && state.integration_methods.contains("scvi")
+      },
+      fromState: [ 
+        "id": "id",
+        "input": "query_processed",
+        "layer": "input_layer",
+        "modality": "modality",
+        "leiden_resolution": "leiden_resolution",
+        "early_stopping": "early_stopping",
+        "early_stopping_monitor": "early_stopping_monitor",
+        "early_stopping_patience": "early_stopping_patience",
+        "early_stopping_min_delta": "early_stopping_min_delta",
+        "max_epochs": "max_epochs",
+        "reduce_lr_on_plateau": "reduce_lr_on_plateau",
+        "lr_factor": "lr_factor",
+        "lr_patience": "lr_patience"
+      ],
+      args: [
+        "obsm_output": "X_scvi_integrated",
+        "obs_batch": "sample_id",
+        "var_input": "filter_with_hvg_query",
+        "uns_neighbors": "scvi_integration_neighbors",
+        "obsp_neighbor_distances": "scvi_integration_neighbor_distances",
+        "obsp_neighbor_connectivities": "scvi_integration_neighbor_connectivities",
+        "obs_cluster": "scvi_integration_leiden",
+        "obsm_umap": "X_scvi_umap"
+      ],
+      toState: [ "query_processed": "output", "scvi_model": "output_model" ]
+    )
+
+    // Annotation methods
+    | celltypist_annotation.run(
+      runIf: { id, state -> state.annotation_methods && state.annotation_methods.contains("celltypist") && state.celltypist_model },
+      fromState: [ 
+        "input": "query_processed",
+        "modality": "modality",
+        "input_var_gene_names": "input_var_gene_names",
+        "input_reference_gene_overlap": "input_reference_gene_overlap",
+        "model": "celltypist_model",
+        "majority_voting": "celltypist_majority_voting"
+      ],
+      args: [
+        // log normalized counts are expected for celltypist
+        "input_layer": "log_normalized",
+        "output_obs_predictions": "celltypist_pred",
+        "output_obs_probability": "celltypist_proba"
+      ],
+      toState: [ "query_processed": "output" ]
+    )
+
+    | celltypist_annotation.run(
+      runIf: { id, state -> state.annotation_methods && state.annotation_methods.contains("celltypist") && !state.celltypist_model },
+      fromState: [
+        "input": "query_processed",
+        "modality": "modality",
+        "input_var_gene_names": "input_var_gene_names",
+        "input_reference_gene_overlap": "input_reference_gene_overlap",
+        "reference": "reference",
+        "reference_layer": "reference_layer_lognormalized_counts",
+        "reference_obs_target": "reference_obs_label",
+        "reference_var_gene_names": "reference_var_gene_names",
+        "reference_obs_batch": "reference_obs_batch",
+        "reference_var_input": "reference_var_input",
+        "feature_selection": "celltypist_feature_selection",
+        "C": "celltypist_C",
+        "max_iter": "celltypist_max_iter",
+        "use_SGD": "celltypist_use_SGD",
+        "min_prop": "celltypist_min_prop",
+        "majority_voting": "celltypist_majority_voting"
+      ],
+      args: [
+        // log normalized counts are expected for celltypist
+        "input_layer": "log_normalized",
+        "output_obs_predictions": "celltypist_pred",
+        "output_obs_probability": "celltypist_proba"
+      ],
+      toState: [ "query_processed": "output" ]
+    )
+
+    | scanvi_scarches_annotation.run(
+      runIf: { id, state -> state.annotation_methods && state.annotation_methods.contains("scanvi_scarches")},
+      fromState: [
+        "id": "id",
+        "input": "query_processed",
+        "modality": "modality",
+        "layer": "input_layer",
+        "input_var_gene_names": "input_var_gene_names",
+        "reference": "reference",
+        "reference_obs_target": "reference_obs_label",
+        "reference_obs_batch_label": "reference_obs_batch",
+        "reference_var_hvg": "reference_var_input",
+        "reference_var_gene_names": "reference_var_gene_names",
+        "unlabeled_category": "reference_obs_label_unlabeled_category",
+        "early_stopping": "early_stopping",
+        "early_stopping_monitor": "early_stopping_monitor",
+        "early_stopping_patience": "early_stopping_patience",
+        "early_stopping_min_delta": "early_stopping_min_delta",
+        "max_epochs": "max_epochs",
+        "reduce_lr_on_plateau": "reduce_lr_on_plateau",
+        "lr_factor": "lr_factor",
+        "lr_patience": "lr_patience",
+        "leiden_resolution": "leiden_resolution",
+        "knn_weights": "knn_weights",
+        "knn_n_neighbors": "knn_n_neighbors"
+      ],
+      args: [
+        "input_obs_batch_label": "sample_id",
+        "output_obs_predictions": "scanvi_knn_pred",
+        "output_obs_probability": "scanvi_knn_proba"
+      ],
+      toState: [ "query_processed": "output" ]
+    )
+
+    | map {id, state ->
+      def new_state = state + ["output": state.query_processed]
+      [id, new_state]
+    }
+
+    | setState(["output", "_meta"])
+
+  emit:
+    output_ch
+}
+
+// inner workflow hook
+def innerWorkflowFactory(args) {
+  return run_wf
+}
+
+// defaults
+meta["defaults"] = [
+  // key to be used to trace the process and determine output names
+  key: null,
+
+  // fixed arguments to be passed to script
+  args: [:],
+
+  // default directives
+  directives: readJsonBlob('''{
+  "tag" : "$id"
+}'''),
+
+  // auto settings
+  auto: readJsonBlob('''{
+  "simplifyInput" : true,
+  "simplifyOutput" : false,
+  "transcript" : false,
+  "publish" : false
+}'''),
+
+  // Apply a map over the incoming tuple
+  // Example: `{ tup -> [ tup[0], [input: tup[1].output] ] + tup.drop(2) }`
+  map: null,
+
+  // Apply a map over the ID element of a tuple (i.e. the first element)
+  // Example: `{ id -> id + "_foo" }`
+  mapId: null,
+
+  // Apply a map over the data element of a tuple (i.e. the second element)
+  // Example: `{ data -> [ input: data.output ] }`
+  mapData: null,
+
+  // Apply a map over the passthrough elements of a tuple (i.e. the tuple excl. the first two elements)
+  // Example: `{ pt -> pt.drop(1) }`
+  mapPassthrough: null,
+
+  // Filter the channel
+  // Example: `{ tup -> tup[0] == "foo" }`
+  filter: null,
+
+  // Choose whether or not to run the component on the tuple if the condition is true.
+  // Otherwise, the tuple will be passed through.
+  // Example: `{ tup -> tup[0] != "skip_this" }`
+  runIf: null,
+
+  // Rename keys in the data field of the tuple (i.e. the second element)
+  // Will likely be deprecated in favour of `fromState`.
+  // Example: `[ "new_key": "old_key" ]`
+  renameKeys: null,
+
+  // Fetch data from the state and pass it to the module without altering the current state.
+  // 
+  // `fromState` should be `null`, `List[String]`, `Map[String, String]` or a function. 
+  // 
+  // - If it is `null`, the state will be passed to the module as is.
+  // - If it is a `List[String]`, the data will be the values of the state at the given keys.
+  // - If it is a `Map[String, String]`, the data will be the values of the state at the given keys, with the keys renamed according to the map.
+  // - If it is a function, the tuple (`[id, state]`) in the channel will be passed to the function, and the result will be used as the data.
+  // 
+  // Example: `{ id, state -> [input: state.fastq_file] }`
+  // Default: `null`
+  fromState: null,
+
+  // Determine how the state should be updated after the module has been run.
+  // 
+  // `toState` should be `null`, `List[String]`, `Map[String, String]` or a function.
+  // 
+  // - If it is `null`, the state will be replaced with the output of the module.
+  // - If it is a `List[String]`, the state will be updated with the values of the data at the given keys.
+  // - If it is a `Map[String, String]`, the state will be updated with the values of the data at the given keys, with the keys renamed according to the map.
+  // - If it is a function, a tuple (`[id, output, state]`) will be passed to the function, and the result will be used as the new state.
+  //
+  // Example: `{ id, output, state -> state + [counts: state.output] }`
+  // Default: `{ id, output, state -> output }`
+  toState: null,
+
+  // Whether or not to print debug messages
+  // Default: `false`
+  debug: false
+]
+
+// initialise default workflow
+meta["workflow"] = workflowFactory([key: meta.config.name], meta.defaults, meta)
+
+// add workflow to environment
+nextflow.script.ScriptMeta.current().addDefinition(meta.workflow)
+
+// anonymous workflow for running this module as a standalone
+workflow {
+  // add id argument if it's not already in the config
+  // TODO: deep copy
+  def newConfig = deepClone(meta.config)
+  def newParams = deepClone(params)
+
+  def argsContainsId = newConfig.allArguments.any{it.plainName == "id"}
+  if (!argsContainsId) {
+    def idArg = [
+      'name': '--id',
+      'required': false,
+      'type': 'string',
+      'description': 'A unique id for every entry.',
+      'multiple': false
+    ]
+    newConfig.arguments.add(0, idArg)
+    newConfig = processConfig(newConfig)
+  }
+  if (!newParams.containsKey("id")) {
+    newParams.id = "run"
+  }
+
+  helpMessage(newConfig)
+
+  channelFromParams(newParams, newConfig)
+    // make sure id is not in the state if id is not in the args
+    | map {id, state ->
+      if (!argsContainsId) {
+        [id, state.findAll{k, v -> k != "id"}]
+      } else {
+        [id, state]
+      }
+    }
+    | meta.workflow.run(
+      auto: [ publish: "state" ]
+    )
+}
+
+// END COMPONENT-SPECIFIC CODE
diff --git a/target/nextflow/single_cell/process_integrate_annotate/nextflow.config b/target/nextflow/single_cell/process_integrate_annotate/nextflow.config
new file mode 100644
index 0000000..ba9ddbc
--- /dev/null
+++ b/target/nextflow/single_cell/process_integrate_annotate/nextflow.config
@@ -0,0 +1,126 @@
+manifest {
+  name = 'single_cell/process_integrate_annotate'
+  mainScript = 'main.nf'
+  nextflowVersion = '!>=20.12.1-edge'
+  version = 'set-public-scope'
+  description = 'A pipeline to process, integrate and annotate single cell (multi-)omics data.\nAvailable integration methods:\n- Harmony\n- scVI\nAvailable annotation methods:\n- CellTypist\n- scANVI (with scArches)\n'
+  author = 'Dorien Roosen, Weiwei Schultz'
+}
+
+process.container = 'nextflow/bash:latest'
+
+// detect tempdir
+tempDir = java.nio.file.Paths.get(
+  System.getenv('NXF_TEMP') ?:
+    System.getenv('VIASH_TEMP') ?: 
+    System.getenv('TEMPDIR') ?: 
+    System.getenv('TMPDIR') ?: 
+    '/tmp'
+).toAbsolutePath()
+
+profiles {
+  no_publish {
+    process {
+      withName: '.*' {
+        publishDir = [
+          enabled: false
+        ]
+      }
+    }
+  }
+  mount_temp {
+    docker.temp            = tempDir
+    podman.temp            = tempDir
+    charliecloud.temp      = tempDir
+  }
+  docker {
+    docker.enabled         = true
+    // docker.userEmulation   = true
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  singularity {
+    singularity.enabled    = true
+    singularity.autoMounts = true
+    docker.enabled         = false
+    podman.enabled         = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  podman {
+    podman.enabled         = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    shifter.enabled        = false
+    charliecloud.enabled   = false
+  }
+  shifter {
+    shifter.enabled        = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    charliecloud.enabled   = false
+  }
+  charliecloud {
+    charliecloud.enabled   = true
+    docker.enabled         = false
+    singularity.enabled    = false
+    podman.enabled         = false
+    shifter.enabled        = false
+  }
+}
+
+process{
+  withLabel: mem1gb { memory = 1000000000.B }
+  withLabel: mem2gb { memory = 2000000000.B }
+  withLabel: mem5gb { memory = 5000000000.B }
+  withLabel: mem10gb { memory = 10000000000.B }
+  withLabel: mem20gb { memory = 20000000000.B }
+  withLabel: mem50gb { memory = 50000000000.B }
+  withLabel: mem100gb { memory = 100000000000.B }
+  withLabel: mem200gb { memory = 200000000000.B }
+  withLabel: mem500gb { memory = 500000000000.B }
+  withLabel: mem1tb { memory = 1000000000000.B }
+  withLabel: mem2tb { memory = 2000000000000.B }
+  withLabel: mem5tb { memory = 5000000000000.B }
+  withLabel: mem10tb { memory = 10000000000000.B }
+  withLabel: mem20tb { memory = 20000000000000.B }
+  withLabel: mem50tb { memory = 50000000000000.B }
+  withLabel: mem100tb { memory = 100000000000000.B }
+  withLabel: mem200tb { memory = 200000000000000.B }
+  withLabel: mem500tb { memory = 500000000000000.B }
+  withLabel: mem1gib { memory = 1073741824.B }
+  withLabel: mem2gib { memory = 2147483648.B }
+  withLabel: mem4gib { memory = 4294967296.B }
+  withLabel: mem8gib { memory = 8589934592.B }
+  withLabel: mem16gib { memory = 17179869184.B }
+  withLabel: mem32gib { memory = 34359738368.B }
+  withLabel: mem64gib { memory = 68719476736.B }
+  withLabel: mem128gib { memory = 137438953472.B }
+  withLabel: mem256gib { memory = 274877906944.B }
+  withLabel: mem512gib { memory = 549755813888.B }
+  withLabel: mem1tib { memory = 1099511627776.B }
+  withLabel: mem2tib { memory = 2199023255552.B }
+  withLabel: mem4tib { memory = 4398046511104.B }
+  withLabel: mem8tib { memory = 8796093022208.B }
+  withLabel: mem16tib { memory = 17592186044416.B }
+  withLabel: mem32tib { memory = 35184372088832.B }
+  withLabel: mem64tib { memory = 70368744177664.B }
+  withLabel: mem128tib { memory = 140737488355328.B }
+  withLabel: mem256tib { memory = 281474976710656.B }
+  withLabel: mem512tib { memory = 562949953421312.B }
+  withLabel: cpu1 { cpus = 1 }
+  withLabel: cpu2 { cpus = 2 }
+  withLabel: cpu5 { cpus = 5 }
+  withLabel: cpu10 { cpus = 10 }
+  withLabel: cpu20 { cpus = 20 }
+  withLabel: cpu50 { cpus = 50 }
+  withLabel: cpu100 { cpus = 100 }
+  withLabel: cpu200 { cpus = 200 }
+  withLabel: cpu500 { cpus = 500 }
+  withLabel: cpu1000 { cpus = 1000 }
+}
+
+includeConfig("nextflow_labels.config")
diff --git a/target/nextflow/single_cell/process_integrate_annotate/nextflow_labels.config b/target/nextflow/single_cell/process_integrate_annotate/nextflow_labels.config
new file mode 100644
index 0000000..8c05307
--- /dev/null
+++ b/target/nextflow/single_cell/process_integrate_annotate/nextflow_labels.config
@@ -0,0 +1,66 @@
+process {
+  // Default resources for components that hardly do any processing
+  memory = { 2.GB * task.attempt }
+  cpus = 1
+
+  // Retry for exit codes that have something to do with memory issues
+  errorStrategy = { task.exitStatus in 137..140 ? 'retry' : 'terminate' }
+  maxRetries = 3
+  maxMemory = null
+
+  // CPU resources
+  withLabel: singlecpu { cpus = 1 }
+  withLabel: lowcpu { cpus = 4 }
+  withLabel: midcpu { cpus = 10 }
+  withLabel: highcpu { cpus = 20 }
+  
+  // Memory resources
+  withLabel: verylowmem { memory = { get_memory( 4.GB * task.attempt ) } }
+  withLabel: lowmem { memory = { get_memory( 8.GB * task.attempt ) } }
+  withLabel: midmem { memory = { get_memory( 16.GB * task.attempt ) } }
+  withLabel: highmem { memory = { get_memory( 64.GB * task.attempt ) } }
+  withLabel: veryhighmem { memory = { get_memory( 75.GB * task.attempt ) } }
+
+  // Disk space
+  withLabel: lowdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: middisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: highdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  withLabel: veryhighdisk {
+    disk = {process.disk ? process.disk : null}
+  }
+  // NOTE: The above labels intentionally do not have an effect by default.
+  // The user should set the disk space requirements by adding the following
+  // to the compute environment:
+  //
+  // withLabel: lowdisk { disk = { 20.GB * task.attempt } }
+  // withLabel: middisk { disk = { 100.GB * task.attempt } }
+  // withLabel: highdisk { disk = { 200.GB * task.attempt } }
+  // withLabel: veryhighdisk { disk = { 500.GB * task.attempt } }
+}
+
+def get_memory(to_compare) {
+  if (!process.containsKey("maxMemory") || !process.maxMemory) {
+    return to_compare
+  }
+
+  try {
+    if (process.containsKey("maxRetries") && process.maxRetries && task.attempt == (process.maxRetries as int)) {
+      return process.maxMemory
+    }
+    else if (to_compare.compareTo(process.maxMemory as nextflow.util.MemoryUnit) == 1) {
+      return max_memory as nextflow.util.MemoryUnit
+    }
+    else {
+      return to_compare
+    }  
+  } catch (all) {
+        println "Error processing memory resources. Please check that process.maxMemory '${process.maxMemory}' and process.maxRetries '${process.maxRetries}' are valid!"
+        System.exit(1)
+  }
+}
diff --git a/target/nextflow/single_cell/process_integrate_annotate/nextflow_schema.json b/target/nextflow/single_cell/process_integrate_annotate/nextflow_schema.json
new file mode 100644
index 0000000..cadc4c2
--- /dev/null
+++ b/target/nextflow/single_cell/process_integrate_annotate/nextflow_schema.json
@@ -0,0 +1,488 @@
+{
+  "$schema": "https://json-schema.org/draft/2020-12/schema",
+  "title": "process_integrate_annotate",
+  "description": "A pipeline to process, integrate and annotate single cell (multi-)omics data.\nAvailable integration methods:\n- Harmony\n- scVI\nAvailable annotation methods:\n- CellTypist\n- scANVI (with scArches)\n",
+  "type": "object",
+  "$defs": {
+    "outputs": {
+      "title": "Outputs",
+      "type": "object",
+      "description": "The output file to write the annotated dataset to.",
+      "properties": {
+        "output": {
+          "type": "string",
+          "format": "path",
+          "description": "The output file.\n",
+          "help_text": "Type: `file`, multiple: `False`, required, default: `\"$id.$key.output.h5mu\"`, direction: `output`, example: `\"output.h5mu\"`. ",
+          "default": "$id.$key.output.h5mu"
+        }
+      }
+    },
+    "input (query) data arguments": {
+      "title": "Input (query) data arguments",
+      "type": "object",
+      "description": "The input query dataset(s) to be annotated",
+      "properties": {
+        "id": {
+          "type": "string",
+          "description": "ID of the sample.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"foo\"`. "
+        },
+        "input": {
+          "type": "string",
+          "format": "path",
+          "exists": true,
+          "description": "Input query dataset(s) to be annotated",
+          "help_text": "Type: `file`, multiple: `False`, required, direction: `input`, example: `\"input.h5mu\"`. "
+        },
+        "modality": {
+          "type": "string",
+          "description": "Modality to be processed",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"rna\"`. ",
+          "default": "rna"
+        },
+        "input_layer": {
+          "type": "string",
+          "description": "The layer in the input data containing the raw counts, if .X is not to be used.",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "input_var_gene_names": {
+          "type": "string",
+          "description": "The name of the adata var column containing gene names; when no gene_name_layer is provided, the var index will be used.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "input_reference_gene_overlap": {
+          "type": "integer",
+          "description": "The minimum number of genes present in both the reference and query datasets.\n",
+          "help_text": "Type: `integer`, multiple: `False`, default: `100`. ",
+          "default": 100
+        }
+      }
+    },
+    "reference data arguments": {
+      "title": "Reference data arguments",
+      "type": "object",
+      "description": "Dataset to be used as a reference for label transfer and to train annotation algorithms on.",
+      "properties": {
+        "reference": {
+          "type": "string",
+          "format": "path",
+          "description": "The reference dataset in .h5mu format to be used as a reference mapper and to train annotation algorithms on.\n",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"reference.h5mu\"`. "
+        },
+        "reference_layer_raw_counts": {
+          "type": "string",
+          "description": "The layer in the reference dataset containing the raw counts, if .X is not to be used.",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "reference_layer_lognormalized_counts": {
+          "type": "string",
+          "description": "The layer in the reference dataset containing the log-normalized counts, if .X is not to be used.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"log_normalized\"`. ",
+          "default": "log_normalized"
+        },
+        "reference_var_gene_names": {
+          "type": "string",
+          "description": "The name of the adata .var column containing gene names if the .var index is not to be used.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "reference_obs_batch": {
+          "type": "string",
+          "description": "The .obs column of the reference dataset containing the batch information.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "reference_obs_label": {
+          "type": "string",
+          "description": "The `.obs` key of the target labels to tranfer.",
+          "help_text": "Type: `string`, multiple: `False`, example: `\"cell_type\"`. "
+        },
+        "reference_obs_label_unlabeled_category": {
+          "type": "string",
+          "description": "Value in the --reference_obs_label field that indicates unlabeled observations",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"Unkown\"`. ",
+          "default": "Unkown"
+        },
+        "reference_var_input": {
+          "type": "string",
+          "description": ".var column containing highly variable genes",
+          "help_text": "Type: `string`, multiple: `False`. "
+        }
+      }
+    },
+    "methods": {
+      "title": "Methods",
+      "type": "object",
+      "description": "The available annotation and integration methods to integrate and/or annotate the query dataset(s) with.",
+      "properties": {
+        "integration_methods": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Integration methods to be executed.",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"harmony;scvi\"]`, choices: ``harmony`, `scvi``. "
+        },
+        "annotation_methods": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Annotation methods to be executed.",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"celltypist;scanvi_scarches\"]`, choices: ``celltypist`, `scanvi_scarches``. "
+        }
+      }
+    },
+    "pre-processing options: rna filtering": {
+      "title": "Pre-processing options: RNA filtering",
+      "type": "object",
+      "description": "Pre-processing options for filtering RNA data",
+      "properties": {
+        "rna_min_counts": {
+          "type": "integer",
+          "description": "Minimum number of counts captured per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "rna_max_counts": {
+          "type": "integer",
+          "description": "Maximum number of counts captured per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `5000000`. "
+        },
+        "rna_min_genes_per_cell": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `200`. "
+        },
+        "rna_max_genes_per_cell": {
+          "type": "integer",
+          "description": "Maximum of non-zero values per cell.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `1500000`. "
+        },
+        "rna_min_cells_per_gene": {
+          "type": "integer",
+          "description": "Minimum of non-zero values per gene.",
+          "help_text": "Type: `integer`, multiple: `False`, example: `3`. "
+        },
+        "rna_min_fraction_mito": {
+          "type": "number",
+          "description": "Minimum fraction of UMIs that are mitochondrial.",
+          "help_text": "Type: `double`, multiple: `False`, example: `0.0`. "
+        },
+        "rna_max_fraction_mito": {
+          "type": "number",
+          "description": "Maximum fraction of UMIs that are mitochondrial.",
+          "help_text": "Type: `double`, multiple: `False`, example: `0.2`. "
+        }
+      }
+    },
+    "pre-processing options: highly variable features detection": {
+      "title": "Pre-processing options: Highly variable features detection",
+      "type": "object",
+      "description": "Pre-processing options for detecting highly variable features",
+      "properties": {
+        "n_hvg": {
+          "type": "integer",
+          "description": "Number of highly-variable features to keep",
+          "help_text": "Type: `integer`, multiple: `False`, default: `2000`. ",
+          "default": 2000
+        }
+      }
+    },
+    "pre-processing options: mitochondrial & ribosomal gene detection": {
+      "title": "Pre-processing options: Mitochondrial & Ribosomal Gene Detection",
+      "type": "object",
+      "description": "Pre-processing options for detecting mitochondrial genes",
+      "properties": {
+        "var_name_mitochondrial_genes": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding the mitochondrial genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "var_name_ribosomal_genes": {
+          "type": "string",
+          "description": "In which .var slot to store a boolean array corresponding the ribosomal genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_name_mitochondrial_fraction": {
+          "type": "string",
+          "description": "When specified, write the fraction of counts originating from mitochondrial genes \n(based on --mitochondrial_gene_regex) to an .obs column with the specified name.\nRequires --var_name_mitochondrial_genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "obs_name_ribosomal_fraction": {
+          "type": "string",
+          "description": "When specified, write the fraction of counts originating from ribosomal genes \n(based on --ribosomal_gene_regex) to an .obs column with the specified name.\nRequires --var_name_ribosomal_genes.\n",
+          "help_text": "Type: `string`, multiple: `False`. "
+        },
+        "mitochondrial_gene_regex": {
+          "type": "string",
+          "description": "Regex string that identifies mitochondrial genes from --var_gene_names.\nBy default will detect human and mouse mitochondrial genes from a gene symbol.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"^[mM][tT]-\"`. ",
+          "default": "^[mM][tT]-"
+        },
+        "ribosomal_gene_regex": {
+          "type": "string",
+          "description": "Regex string that identifies ribosomal genes from --var_gene_names.\nBy default will detect human and mouse ribosomal genes from a gene symbol.\n",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"^[Mm]?[Rr][Pp][LlSs]\"`. ",
+          "default": "^[Mm]?[Rr][Pp][LlSs]"
+        }
+      }
+    },
+    "pre-processing options: qc metrics calculation options": {
+      "title": "Pre-processing options: QC metrics calculation options",
+      "type": "object",
+      "description": "Pre-processing options for calculating QC metrics",
+      "properties": {
+        "var_qc_metrics": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "Keys to select a boolean (containing only True or False) column from .var.\nFor each cell, calculate the proportion of total values for genes which are labeled 'True', \ncompared to the total sum of the values for all genes",
+          "help_text": "Type: `string`, multiple: `True`, example: `[\"ercc,highly_variable\"]`. "
+        }
+      }
+    },
+    "harmony integration options": {
+      "title": "Harmony integration options",
+      "type": "object",
+      "description": "Specifications for harmony integration.",
+      "properties": {
+        "harmony_theta": {
+          "type": "array",
+          "items": {
+            "type": "number"
+          },
+          "description": "Diversity clustering penalty parameter",
+          "help_text": "Type: `double`, multiple: `True`, default: `[2.0]`, example: `[0.0;1.0;2.0]`. ",
+          "default": [
+            2.0
+          ]
+        },
+        "harmony_obs_covariates": {
+          "type": "array",
+          "items": {
+            "type": "string"
+          },
+          "description": "The .obs field(s) that define the covariate(s) to regress out.",
+          "help_text": "Type: `string`, multiple: `True`, required, default: `[\"sample_id\"]`, example: `[\"batch\";\"sample\"]`. ",
+          "default": [
+            "sample_id"
+          ]
+        }
+      }
+    },
+    "scvi, scanvi and scarches training options": {
+      "title": "scVI, scANVI and scArches training options",
+      "type": "object",
+      "description": "Training arguments for scVI, scANVI and scArches. Relevant for --annotation_methods 'scvi_knn' and 'scanvi_scarches'.",
+      "properties": {
+        "early_stopping": {
+          "type": "boolean",
+          "description": "Whether to perform early stopping with respect to the validation set.",
+          "help_text": "Type: `boolean`, multiple: `False`. "
+        },
+        "early_stopping_monitor": {
+          "type": "string",
+          "description": "Metric logged during validation set epoch.",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"elbo_validation\"`, choices: ``elbo_validation`, `reconstruction_loss_validation`, `kl_local_validation``. ",
+          "enum": [
+            "elbo_validation",
+            "reconstruction_loss_validation",
+            "kl_local_validation"
+          ],
+          "default": "elbo_validation"
+        },
+        "early_stopping_patience": {
+          "type": "integer",
+          "description": "Number of validation epochs with no improvement after which training will be stopped.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `45`. ",
+          "default": 45
+        },
+        "early_stopping_min_delta": {
+          "type": "number",
+          "description": "Minimum change in the monitored quantity to qualify as an improvement, i.e",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.0`. ",
+          "default": 0.0
+        },
+        "max_epochs": {
+          "type": "integer",
+          "description": "Number of passes through the dataset, defaults to (20000 / number of cells) * 400 or 400; whichever is smallest.",
+          "help_text": "Type: `integer`, multiple: `False`. "
+        },
+        "reduce_lr_on_plateau": {
+          "type": "boolean",
+          "description": "Whether to monitor validation loss and reduce learning rate when validation set `lr_scheduler_metric` plateaus.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `true`. ",
+          "default": true
+        },
+        "lr_factor": {
+          "type": "number",
+          "description": "Factor to reduce learning rate.",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.6`. ",
+          "default": 0.6
+        },
+        "lr_patience": {
+          "type": "number",
+          "description": "Number of epochs with no improvement after which learning rate will be reduced.",
+          "help_text": "Type: `double`, multiple: `False`, default: `30.0`. ",
+          "default": 30.0
+        }
+      }
+    },
+    "celltypist reference model": {
+      "title": "CellTypist reference model",
+      "type": "object",
+      "description": "The CellTypist reference model to use for annotation. If not provided, the reference dataset will be used for model training.",
+      "properties": {
+        "celltypist_model": {
+          "type": "string",
+          "format": "path",
+          "description": "Pretrained model in pkl format",
+          "help_text": "Type: `file`, multiple: `False`, direction: `input`, example: `\"pretrained_model.pkl\"`. "
+        }
+      }
+    },
+    "celltypist annotation options": {
+      "title": "CellTypist annotation options",
+      "type": "object",
+      "description": "Specifications for CellTypist annotation.",
+      "properties": {
+        "celltypist_feature_selection": {
+          "type": "boolean",
+          "description": "Whether to perform feature selection.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "celltypist_majority_voting": {
+          "type": "boolean",
+          "description": "Whether to refine the predicted labels by running the majority voting classifier after over-clustering.",
+          "help_text": "Type: `boolean`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "celltypist_C": {
+          "type": "number",
+          "description": "Inverse of regularization strength in logistic regression.",
+          "help_text": "Type: `double`, multiple: `False`, default: `1.0`. ",
+          "default": 1.0
+        },
+        "celltypist_max_iter": {
+          "type": "integer",
+          "description": "Maximum number of iterations before reaching the minimum of the cost function.",
+          "help_text": "Type: `integer`, multiple: `False`, default: `1000`. ",
+          "default": 1000
+        },
+        "celltypist_use_SGD": {
+          "type": "boolean",
+          "description": "Whether to use the stochastic gradient descent algorithm.",
+          "help_text": "Type: `boolean_true`, multiple: `False`, default: `false`. ",
+          "default": false
+        },
+        "celltypist_min_prop": {
+          "type": "number",
+          "description": "\"For the dominant cell type within a subcluster, the minimum proportion of cells required to \nsupport naming of the subcluster by this cell type",
+          "help_text": "Type: `double`, multiple: `False`, default: `0.0`. ",
+          "default": 0.0
+        }
+      }
+    },
+    "clustering options": {
+      "title": "Clustering options",
+      "type": "object",
+      "description": "Arguments for Leiden clustering. Only relevant for --annotation_methods `scvi_knn`, `scanvi_scarches` and `harmony_knn`.",
+      "properties": {
+        "leiden_resolution": {
+          "type": "array",
+          "items": {
+            "type": "number"
+          },
+          "description": "Control the coarseness of the clustering",
+          "help_text": "Type: `double`, multiple: `True`, default: `[1.0]`. ",
+          "default": [
+            1.0
+          ]
+        }
+      }
+    },
+    "neighbor classifier arguments": {
+      "title": "Neighbor classifier arguments",
+      "type": "object",
+      "description": "Arguments related to calculating the n nearest neighbors. Only relevant for --annotation_methods `scvi_knn`, `scanvi_scarches` and `harmony_knn`.",
+      "properties": {
+        "knn_weights": {
+          "type": "string",
+          "description": "Weight function used in prediction",
+          "help_text": "Type: `string`, multiple: `False`, default: `\"uniform\"`, choices: ``uniform`, `distance``. ",
+          "enum": [
+            "uniform",
+            "distance"
+          ],
+          "default": "uniform"
+        },
+        "knn_n_neighbors": {
+          "type": "integer",
+          "description": "The number of neighbors to use in k-neighbor graph structure used for fast approximate nearest neighbor search with PyNNDescent",
+          "help_text": "Type: `integer`, multiple: `False`, default: `15`. ",
+          "default": 15
+        }
+      }
+    },
+    "nextflow input-output arguments": {
+      "title": "Nextflow input-output arguments",
+      "type": "object",
+      "description": "Input/output parameters for Nextflow itself. Please note that both publishDir and publish_dir are supported but at least one has to be configured.",
+      "properties": {
+        "publish_dir": {
+          "type": "string",
+          "description": "Path to an output directory.",
+          "help_text": "Type: `string`, multiple: `False`, required, example: `\"output/\"`. "
+        }
+      }
+    }
+  },
+  "allOf": [
+    {
+      "$ref": "#/$defs/outputs"
+    },
+    {
+      "$ref": "#/$defs/input (query) data arguments"
+    },
+    {
+      "$ref": "#/$defs/reference data arguments"
+    },
+    {
+      "$ref": "#/$defs/methods"
+    },
+    {
+      "$ref": "#/$defs/pre-processing options: rna filtering"
+    },
+    {
+      "$ref": "#/$defs/pre-processing options: highly variable features detection"
+    },
+    {
+      "$ref": "#/$defs/pre-processing options: mitochondrial & ribosomal gene detection"
+    },
+    {
+      "$ref": "#/$defs/pre-processing options: qc metrics calculation options"
+    },
+    {
+      "$ref": "#/$defs/harmony integration options"
+    },
+    {
+      "$ref": "#/$defs/scvi, scanvi and scarches training options"
+    },
+    {
+      "$ref": "#/$defs/celltypist reference model"
+    },
+    {
+      "$ref": "#/$defs/celltypist annotation options"
+    },
+    {
+      "$ref": "#/$defs/clustering options"
+    },
+    {
+      "$ref": "#/$defs/neighbor classifier arguments"
+    },
+    {
+      "$ref": "#/$defs/nextflow input-output arguments"
+    }
+  ]
+}